Skip to main content

Full text of "The diamond mines of South Africa; some account of their rise and development"

See other formats


UNIVERSITY OF CALIFORNIA. 



O1KT OK 

^. r r 



- . 



.3 




^, > 

-<**V>" 



555" 



i 



> i. 









/ ^ 

^ 







- 




\ 

i 



mm 



I . 

















I 








^ 



^ 
















* 







\ 




/ 



< X 



r 




t?/~ 




THE DIAMOND MINES OF 
SOUTH AFRICA 

SOME ACCOUNT OF THEIR RISE 
AND DEVELOPMENT 



THE DIAMOND MINES OF 
SOUTH AFRICA 



SOME ACCOUNT OF THEIR RISE 
AND DEVELOPMENT 



BY 



GARDNER F. WILLIAMS, M.A. 

GENERAL MANAGER OF DE BEERS CONSOLIDATED MINES, LTD. 



ILLUSTRATED 




Nefa gorfc 
THE MACMILLAN COMPANY 

LONDON : MACMILLAN & CO., LTD. 
I9O2 

All rights reserved 



EDITION DE LUXE. 

Of this Edition only One Hundred Copies have been 
printed. 

Tbi, Cofy i, Jfo~ 



COPYRIGHT, 1901, 
BY THE MACMILLAN COMPANY. 



Set up and electrotyped September, 1902. 




NortooolJ 

J. 5. Cuibing & Co. Berwick (3 Smith 

Norwood, Man., U.S.A. 



Contents 



CHAPTER PAGE 

I. The Ancient Adamas ........ i 

II. In Traditional Ophir Land . . . . . . .32 

III. The Pioneer Advance . . . . . . . -87 

IV. The Discovery . . . . . . . . .114 

V. The Camps on the Vaal . . . . . . .140 

VI. The Rush to Kimberley 164 

VII. The Great White Camps 190 

VIII. Opening the Craters . . . . . . . .220 

IX. The Moving Men ........ 267 

X. The Essential Combination . . . . . . .297 

XI. Systematic Mining ........ 307 

XII. Winning the Diamonds . . . . . . .360 

XIII. Obstacles and Perils 384 

XIV. The Workers in the Mines 407 

XV. The Mining Towns 450 

XVI. Formation of the Diamond ....... 479 

XVII. The Diamond Market 511 

XVIII. Cutting and Polishing 528 

XIX. An Uplifting Power . . . . . . . 552 

APPENDIX 

I. The Mines besieged . . . . . . . .605 

II. Winding Engines for the Main Shaft, Kimberley Mine . . 667 

III. Report on Pumping Plant for Kimberley Mine .... 669 

IV. Relative Value of Coals ....... 670 

V. Statistics of De Beers Company . . . . . .671 



Illustrations 



PAGE 

The Koh-i-nur (Old Cutting) I 

i. A Black Diamond in Gold Setting. 2. Ordinary Window Glass. 

3. A Pink Diamond ........ 2 

The Shah 3 

The Egyptian Pascha ......... 4 

The Polar Star 8 

The Hope Blue 8 

The Empress Eugenie . . . . . . . . .15 

The Nassak 16 

The Great Mogul 17 

The Sancy ........... 25 

The Koh-i-nur (Present Cutting) . . . . . . -27 

The Orloff 28 

The Regent ........... 29 

The Florentine .......... 30 

The Piggott . . . . . . . . . .30 

The Star of the South ......... 30 

Dutch Ships of the Eighteenth Century ..... 40, 44, 45 

Dutch Ships of the Seventeenth Century . . . . . 41, 42, 43 

Insiza Ruins .......... 48, 49, 50 

Khami Ruins ......... 51, 52, 53, 54 

Gold Ornaments found in Ancient Ruins . . . . . .52 

Zimbabwe Ruins . . . . . . 55, 56, 57, 58, 59, 60 

The Old East India House, Leadenhall Street, London . . . .61 

The Landing of Van Riebeeck . . . . . . . .62 

Portrait of Johan Antonyse van Riebeeck ...... 63 

Portrait of Maria de la Querellerie . . . . . . .63 

Constantia ........... 64 

Vergelegen ........... 69 

Boschendal 71, 72 

Entrance to Boschendal . . . . . . . . .72 

Botanic Gardens .......... 73 

Lekkerwijn ........ , , . 74 

vii 



viii ILLUSTRATIONS 

PAGE 

Bien Donne, Groot Constantia . . . . . . . 75, 76 

Overmantel and Old Dutch Relics . . . . . . -75 

Farm House, Klein Drakenstein . . . . . . . . -7 

Palmeit Vallei 77, 80 

A Wine Farm at Klein Drakenstein . . . . . . .78 

Muller's Farm, Achter Paarl ....... 78, 79, 80 

Mooi Kelder, Lower Paarl . . . . . . . .81 

Plaisis de Merle, Groot Drakenstein . . . . . . .81 

Donkerhoek, Groot Drakenstein . . . . . . . .82 

A Wine Cellar. Herd of Cape Goats 82 

Tatr, 1757 . .83 

An Old Farm House, Lower Paarl ....... 84 

Farm House, Achter Paarl ........ 84 

Brand Solder 85 

Cape Cart 85, 276 

The Gate of the Castle 86 

Zulu Chief Cetawayo and Part of his Family . . . . .92 

Zulu Prince Dinizulu . . . . . . . . -93 

Zulu Family ........... 93 

A Zulu and his Ten Wives ........ 94 

Zulu Kraal and Huts ......... 95 

Zulu Hut in course of Construction .....'.. 96 

Zulu Woman grinding Corn ........ 97 

Zulu Women .......... 98 

Zulus smoking Indian Hemp ........ 99 

Old Zulu Women taking Kafir Beer to a Wedding .... 99 

Zulu Girls . . . . . . . . . . . .100 

Native Laborers in War Dress . . . . . <" . 101 

Trekbok (Springbok) Hunting . . . . . . . .102 

Zulu in War Dress . . . . . . . . . .103 

Zulu Jim Cameel ......... 105 

A Zulu Laborer in War Attire . . . . . . . .108 

Nest of Social Grosbeak . . . . . . . . i'i 2 

John O'Reilly 120 

Mr. Lorenzo Boyes . . . . . . . . .121 

Dr. W. Guybon Atherstone . . . . . . . .121 

Pniel Diggings . . . . . . . . . .139 

Delport's Hope, Vaal River Diggings . . . . . . .142 

Diggers' Camps on the Vaal River . . . . . . .143 

River Diggings at Gong Gong, 1880 . . . . . . .147 

Vaal River Diggings . , , . , , , . . 149 



ILLUSTRATIONS ix 

PAGE 

River Diggings, Waldek Plant . . . . . . . .151 

Pniel Diggings, Vaal River . . . . . . . .152 

Klip-drift, Early River Diggings . . . . . . .152 

Gong Gong . . . . . . . . . . 153 

Washing Diamond Gravel by Machinery at Gong Gong, 1880 . . 155 

Lightning at Kimberley . . . . . . . . . 1 56 

Day View, Same Scene . . . . . . . . .156 

Largest River Diamond ever found in South Africa . . . . .158 

Views of Klip-drift .......... 162 

Kimberley, 1872 .......... 169 

Mrs. Rawstorne . . . . . . . . . .173 

Mr. T. B. Kisch 174 

Kimberley Mine just after the Discovery, July, 1871 . . . 175, 176 
Fleetwood Rawstorne . . . . . . . . .177 

Native Chiefs . . . . . . . . . .179 

The First Government House and Buildings of the Colony of Griqualand 

West 1 80 

Sir Richard Southey's Residence, Kimberley . . . . .181 

Stephen J. Paul Kriiger . . . . . . . . .185 

Coach leaving Kimberley for the Coast, 1875 .187 

Kimberiey, before the Discovery of Diamonds . . . . .190 

Dutoitspan . . . . . . . . . . 191, 192 

Kimberley, 1871 . . ~f~" ....... 193 

Kimberley Mine, 1871 . . . . . . . . .194 

Around Kimberley Mine, 1871. . . . . . . -'95 

Kimberley Mine, 1872 197, 203, 205, 208, 209 

Centre Block, Kimberley Mine, 1874 200 

The Roadways, Kimberley Mine, 1871-1872 20 z 

Roads in Kimberley Mine, 1871-1872 204,207 

Market Square, Kimberley . . . . . . .212, 213 

Natives resting, on their Way to the Mines . . . . z 1 8 

The Breaking up of the Roads, Kimberley Mine, 1872 . . . .221 

Miners going to Work ......... 222 

The Hand Drums used for Winding-up the Blue Ground . . . 223 

De Beers Mine, 1873 223 

Kimberley Mine, 1873 - 2Z 4 

Kimberley Mine, 1874 225 

Natives carrying Ground out of Dutoitspan in Buckets . . . .226 
Back View of the Staging with Grooved Wheels, at Kimberley . . . 226 

Kimberley Mine, 1875 - . . . 227 

Snow in Kimberley Mine, June 2 1 , 1876 . . . . . .228 



x ILLUSTRATIONS 

PAGE 

Method of Hauling, De Beers Mine, 1873 228 

The First Horse Whim, Kimberley Mine, 1874 . ... 229 

Hauling Gear and Jumpers, Kimberley Mine, 1878 . . . :g 

A Nook in Kimberley Mine, 1874 230 

The Horse Whims, Kimberley Mine, 1875 231 

Hauling Gear, Dutoitspan Mine, 1876 ...... 232 

Surface Loading Boxes ...... 233 

Aerial Trams and Surface Chutes, De Beers Mine, 1885 . . .233 

Hauling Gear, Kimberley Mine, 1885 234 

The French Company's Sling Gear, 1885 235 

Loading Tubs at Bottom of Kimberley Mine, 1885 . . . . 236 

The Standard Company's Claim, Bottom of Kimberley Mine, 1885 . . 237 
Bottom of Kimberley Open Workings . . . . . . .238 

Pumping Engine in Kimberley Mine, 1875 ...... 239 

Incline Tramway for Hauling Reef, 1878 . . . . . . 240 

Hauling Reef, Kimberley Mine, 1873 ...... 241 

Reef Falls, Kimberley Mine, 1 88 1 242 

Steam Pumping Engine, De Beers Mine, 1879 . . . . . 243 
The Central Company's Shaft, Kimberley Mine, 1885 . . . . 244 

The Bottom of Kimberley Mine, 1885 245 

Plan of Kimberley Mine, 1883 . . . . . . . .246 

Reef Slips, Kimberley Mine, 1874 . . 247 

Kimberley Mine, showing how the Ground cracked before Subsidence . 247 

The Central Company's Atkins Shaft ....... 248 

The Last of Open Working, Kimberley Mine, 1889 .... 249 

R. D. Atkins 250 

No. 2 Incline Shaft, De Beers Mine . . . . . . .251 

Eldorado Road, Dutoitspan Mine, 1874 . . . . . 251 

Claims in Dutoitspan Mine . . . . . . . .252 

Bultfontein Mine, 1879 2 S3 

The First Rotary Washing Machine 254 

Another Early Washing Machine, 1874 . . . . . .255 

Horse-power Washing Machine, 1875 . . . . . . , 255 

Early Horse-power Washing Machine, 1874 ..... 256 

The First Washing Machine with Elevator to carry away the Tailings . 257 

Washing Gear, Bultfontein Mine . . . . . . .258 

Steam Washing Gear, Kimberley Mine . . . . . .259 

Webb's Washing Machine, 1878 260 

Cape of Good Hope Company's Washing Gear, 1878 . . . .261 

Washing Gear, Dutoitspan Mine ....... 262 

Washing Gear, Bultfontein Mine, 1878 ,,,,,. 263 



ILLUSTRATIONS x i 

PAGE 

Mr. Barney Barnato ....... 268 

C. J. Rhodes, when a Student at Oxford . . . . . .272 

Cape Town ........... 273 

Silver Trees . . . . . . . . . . . 27C 

Mr. C. D. Rudd 279 

Mr. Robert English ......... 270 

Plan of Kimberley Mine, 1882 ........ 282 

House of Parliament, Cape Town ....... 284 

Avenue of Oaks, Cape Town . . . . . . . .285 

Mr. Carl Meyer .......... 287 

Mr. Alfred Beit 289 

The Diamond Market, Kimberley, 1875 . . . . . . 290 

The Right Honorable Cecil John Rhodes, and Alfred Beit, Esq., October, 

1901 292 

Fac-simile of Check given in Payment for Kimberley Mine . . . 295 
A Group of Directors, De Beers Mines ...... 300 

Group of Life Governors, Directors, Manager, and Secretary, De Beers Mine 303 
Mr. E. R. Tymms ......... 306 

The Last of Open Mining, Kimberley . . . . . .308 

Plan of De Beers Mine ....... 309, 316, 318 

Section through De Beers Mine ....... 310, 311 

Plan of Kimberley Mine . . . . . . . . .312 

Section of Kimberley Mine . . . . . . . -313 

Sketch of Premier Mine . . . . . . . . .318 

Sloping 319, 320 

Timbering Tunnels . . . . . . . . .321, 322 

Natives drilling, De Beers Mine . . . . . . .322 

A Shaft Station . . . . . . . . . .324 

Loading the Trucks . . . . . . . . . .325 

Loading Chutes for Rock Shaft . . . ... . . .326 

Plan of Skip for Six Loads . . . . . . . .327 

Main Shaft, Kimberley Mine . . . . . . . .328 

The Rock Shaft, De Beers Mine 329 

Vertical Tandem Compound-condensing Winding Engines . . 330, 331 
Winding Engine, Kimberley Mine . . . . . . 332 

Mr. Louis I. Seymour . . . . . . . . 33 2 

Plan of Bultfontein Mine . . . . . . . . -333 

Mount Ararat before Blasting ........ 342 

Shots fired 343 

A Second after Firing . . . . . . . . '344 

The Mine filled with Smoke . , . . . . . -345 



xii ILLUSTRATIONS 

PAGE 

After the Smoke has cleared away ....... 346 

Premier Mine, Open Workings .... . . . 34,8 

Premier Mine 350, 351, 353^ 

One of the Early Washing Machines . . . . . . .354 

One of the Present Washing Plants . . . . . . -355 

No. i Washing Plant, De Beers Floors . . . . . 356 

No. 2 Washing Plant, De Beers Floors . . . . 357, 370 

Excelsior Diamond ......... 358, 359 

De Beers Mine and Floors . . . . . . . .361 

Details of zo-Cubic-feet Truck ........ 362 

Details of 1 6-Cubic-feet Truck ........ 363 

De Beers Floors ........ 364, 365, 366 

Mechanical Haulage, Kimberley Mine . . . . . .364 

Kimberley Floors . . . . . . . . . .366 

Harrowing Blue Ground . . . . . . . . .367 

Traction Engine for harrowing Blue Ground . . . . .368 

Loading Pulverized Blue Ground, De Beers Floors . . . . ^ 369 

Washing Machine, De Beers Mine . . . . . . .370 

De Beers Crushing Mill 371.374 

Details of Rotary Washing Machine . . . . . -372.373 

The Pulsator 376 

Sorting Gravel for Diamonds . . . . . ... -377 

Automatic Diamond Sorter .. . . . . . -379 

Mr. James Stewart . . . . . . . . . .381 

^60,000 Parcel of Diamonds ........ 382 

A Day's Diamond Wash ......... 383 

Mr. Lindsay, killed in Fire . . . . . . . .388 

Fire in De Beers Mine, July ii, 1888 389 

No. z Incline Shaft, after the Fire . . . . . 39 2 393 

Waiting for News from the Mine ....... 395 

Men escaping through Tunnel ........ 398 

The Survivors coming up the Terraces . . . . . .401 

General Manager of De Beers Mines and his Staff . . . . . 406 

J. M. Jones ........... 407 

The Engineers, Mechanics, and Workmen who built De Beers Crushing 

Plant 408 

Frank Mandy .......... 409 

G. Scott 409 

Rinderpest ........... 410 

The Game of the Country . . . . . . . .411 

Chiefs of the Batlapin Tribe . . . , . . . .412 



ILLUSTRATIONS xiii 

PAGE 

De Beers Compound . ... 414,415,417,418,419 

De Beers Compound Musical Band of Natives ..... 420 

A Fireside Gathering, Kimberley Compound . . . . .421 

Natives making Coffee, Kimberley Compound . . . . .422 

Open Mine Workers, Kimberley Compound ..... 423 

Kimberley Mine Compound ........ 424 

Natives drilling in the Open Mine . . . . . . -425 

The Last of Open Mining in Dutoitspan Mine ..... 425 

Natives drilling Underground ........ 426 

The Midday Meal 427 

Zulu Workmen, Dutoitspan Mine . . . . . . .429 

Native making Bangles ......... 430 

Mr. Rouliot and Native Workmen . . . . . . . 43 1 

'Mshangaan in War Attire . . . . . . , .433 

Native War Dance .......... 434 

'Mshangaans in War Paint . . . . . . . .435 

Sir Alfred Milner's Visit to Kimberley Compound . . . . .437 

A Quiet Game of Cards . . . . . . . . -438 

Natives playing Chuba ......... 439 

Natives playing Mancala . . . . . . . . .439 

Swimming Bath, De Beers Compound ...... 440 

Natives smoking Indian Hemp . . . . . . . .441 

Diamonds swallowed by a Native ...... 444, 445 

Diamond Thieves ........ 446, 447, 448 

De Beers Machine Shops ......... 449 

Snow in Kimberley, 1876 . . . . . . -45' 

Theatre Royal ......... 452, 453 

The Town Hall 454 

Dutoitspan Road, Kimberley . . . . . . . .454 

Kimberley Hospital . . . . . . . . . -455 

The Kimberley Club 455, 456 

Horns of South African Antelope . . . 457, 458, 459, 460, 461 

Professor J. G. Lawn ......... 462 

Professor Orr ........... 462 

Sir Alfred Milner passing the Offices of De Beers Mines .... 463 

De Beers Offices decorated in Honor of Governor's Visit . . .463 

Kimberley Public Library ......... 464 

The Sanatorium .......... 464 

Masonic Temple .......... 465 

The Post-office, Kimberley ........ 466 

Nazareth House, Kimberley ........ 466 



xiv ILLUSTRATIONS 

PAGE 

The Author's House at Kimberley . . . . . . . -67 

" Smell my Flowers " 468 

Kimberley Race Course . . . . . . . . .468 

The Grand Stand 469 

Bachelors' Quarters, Kenilworth Village . . . . . . 47 1 

Passenger Train for Workmen . . . . . . . .472 

The Road to Kenilworth and Kenilworth Reservoir . . . .472 

Kenilworth Club-house . . . . . . . . .473 

Kenilworth Village ........ 474, 475, 476 

Preparing Trenches for planting Vines and Trees . . . . .477 

Section of De Beers and Kimberley Rock Shafts ..... 480 

De Beers Diamond . . . . . . . . .482 

The Largest Diamond ever found in the Kimberley Mine . . .483 

Fossil Fish from Premier Mine ........ 486 

Irregular Crystallization of Diamonds ...... 488, 489 

Crystal of Diamond . . . . . . . . ( . 490 

Twin Crystal Formations . . . . . ... . .491 

Diamonds of Regular Forms . . . . . . . .493 

Diamonds of Irregular Forms ....... 496, 504 

A Microscopical Diamond . . . . . . . .501 

Smooth Surface of Diamond dissected by Combustion ... . . 503 

Two Views of the Face of a Rough Diamond as seen through the Microscope 505 
Diamond bearing a Smaller Crystal in its Centre ..... 507 

Diamond Sorters and Valuators . . . . . . . .512 

A De Beers Group . . . . . . . . . .515 

The Officials who manage the Benefit Society, De Beers Mines . . 518 

A Group of Officials of De Beers Consolidated Mines Limited . . 519 

Photograph of a Book with the Leaves cut out in the Centre, used by an 

Illicit Diamond Dealer to send Diamonds to England . . .522 
Examination of the Diamonds before and after Cutting . . . .529 

Cleaving ........... 540 

Cutting . . . . 541, 543 

Polishing . . . 54.5, 549 

A Group of Directors, General Manager, and Secretary of De Beers Mines . 552 
Natives riding Bullocks . . . . . . . . -553 

Pioneers Trekking . . . . . . . . . .554 

A Prospecting Expedition to Mashonaland . . . . . .554 

Trekking on the Veld 555 

The first and only Load of Cotton raised in South Africa . . . .556 

Zebra 557 

Coach leaving Kimberley for the Transvaal before the Day of the Railway . 558 



ILLUSTRATIONS X v 



PAGE 



Cape Town and Table Mountain . . . . . . 558, 561 

A Group of Well-known Kimberley Men . . . . . -559 

Eland ............ 560 

Table Mountain, from the West . . . . . . .562 

Lord Milner ........... 563 

Donkey Transport .......... 564 

A South African Farm ......... 565 

Natives shearing Sheep ......... 566 

Dutch Farm in the Karoo ......... 567 

Beaufort West .......... 568 

Norvals Pont Bridge . . . . . . . . .569 

Sir Walter Hely Hutchinsen . . . . . . . .571 

Entrance Groote Schuur . . . . . . . -575 

Groote Schuur after its Destruction by Fire . . . . . -577 

Pioneers Trekking in Mashonaland . . . . . . -579 

Khama, a Noted Chief . . . . . . . . .580 

Khama's Town . . . . . . . . . .581 

Khama's Hut .......... 582 

Matabele Women carrying Water . . . . . . -583 

A Group of Visitors at Groote Schuur. . . . . . .584 

Great Kafir Kraal, Mochudi, Matabeleland . . . . . .585 

Groote Schuur .......... 587 

Major Wilson's Grave ......... 588 

Victoria Falls, Zambesi River ...... 590, 591, 593 

Zambesi River One Mile below Victoria Falls . . . . .594 

Summer House at Groote Schuur ....... 596 

" Stoep," Groote Schuur. ........ 597 

A Corner in Mr. Rhodes's Library ....... 599 

Another View of Mr. Rhodes's Library ...... 600 

A Boer Commando .......... 605 

Lieut. Col. Kekewich and Staff of Imperial Officers .... 606 

Fort on Tailings Heap, Kimberley Mine Floors ..... 607 

One of the Redoubts ......... 608 

Fort Rhodes, Kenilworth, and its Defence Force ..... 609 

View from the Conning Tower ........ 609 

The Waterworks Company's Reservoir as a Fort . . . . .610 

Fort Rhodes, on Top of No. 2 Tailings Heap, De Beers Floors . .610 
No. 2 Redoubt, near Kimberley Mine . . . . . .611 

A Barrier on the Road leading to Kenilworth . . . . .611 

The Sanatorium in Time of Peace . . . . . . .612 

The Sanatorium during the Siege . . . . . . .612 



xvi ILLUSTRATIONS 

PAGE 

Mounted Camp, Kimberley . . . . . . . .613 

Camp of the Royal North Lancashires . . . . . . .614 

Officers of the Diamond Fields Horse . . . . . . .615 

The Diamond Fields Horse at Kenilworth, during Siege of Kimberley . 615 

A Group of the Town Guard . . . . . . . .616 

Defence Guns and Maxims massed in the Gardens . . . .616 

Headquarters Staff, Cape Mounted Police . . . . . .617 

Kimberley Waterworks Reservoir, with Royal Artillery . . . .618 

Premier Mine Fort, Royal Artillery in Action . . . . .618 

Premier Mine Searchlight . . . . . . . . .619 

Canvas House erected for Protection from the Sun and Thunderstorms . 619 
Railway Bridge over the Vaal River at Fourteen Streams .... 620 

Boer Laager, near Kimberley . . . . . . . .621 

Group of Typical Boers . . . . . . . . .621 

A Group of Mercenaries fighting with the Boers . . . / . . 622 

Code Dispatches received during the Siege . . . . . .623 

Typical Boer ........... 624 

Major W. H. E. Murray ........ 624 

Armored Locomotive . . . . . . . . .625 

Armored Train . . . . . . . . . .625 

Railway Bridge at Modder River ...'.... 626 

Effect of a Nine-Pound Shell . . . . . . . .627 

Trophies of the Siege ......... 628 

Boer Nine-Pound Shrapnel . . . . . . . . 629 

British and Boer Shells fired at Kimberley ...... 630 

Site of Cronje's Laager, Magersfontein . . . . . .631 

Conning Tower, De Beers No. I Shaft . . . ^" 632 
The Funeral of Colonel Scott-Turner and the Men who fell with him . 633 

Boer Trenches at Magersfontein ........ 634 

Plan through Tomb .......... 634 

The Honoured Dead Memorial . . . . . . . .635 

Plan through Columns . . . . . . . . .-635 

"Long Cecil" as a Mild Steel Billet - . 636 

" Long Cecil " in course of Construction . . . . . .637 

"Long Cecil " just before it was taken out of the Workshops . . .638 
Construction of " Long Cecil " . . . . . . . .639 

Boring and Rifling Tools ......... 640 

Shell for 4.1 inch B. L. Siege Gun and for 2.5 inch Gun . . .641 
"Long Cecil" and the De Beers Men who made it .... 642 

"Long Cecil" 643 

"Long Cecil" firing at the Boers on Carter's Ridge . . . .643 



ILLUSTRATIONS xvii 

PAGE 

"Long Cecil," made in Kimberley, and Royal Artillery Gun sent to defend 

Kimberley .......... 644 

Casting Shells for Seven- Pounders, De Beers Foundry .... 645 

The Soup Kitchen .......... 645 

Comparison of Shells ......... 646 

Notice on the Market Buildings during the Siege ..... 647 

Inhabitants of Kimberley waiting for their Daily Allowance of Horse Meat . 648 
Band of the Kimberley Volunteer Regiment playing at the Mounted Camp 

during the Siege ......... 649 

The Mafeking "Long Tom" 650 

Boer Shells . . . . . . . . . . .651 

Premier Studio, showing Effect of 100- Pound Shell . . . .652 

Effect of a loo-Pounder ........ 653, 654 

Mr. Compton's Drawing-room barricaded for Shelter from Boer Guns . 654 
Shell-proof constructed by the Public Works Department . . . .6;; 

Excavations in the Tailing Heaps at Beaconsfield, used as Shelters . . 656 
Shell-proof, after the Siege . . . . . . . .656 

Shell-proof at the Convent . . . . . . . .657 

Shell-proof Dugouts . ......... 657 

Notice sent round the Town during the Shelling by the Boers' 100- Pound 

Gun ........... 658 

Shelters for Women and Children during firing of loo-Pound Boer Gun . 659 
Too Late ! Two Siege Guns arrived after the Siege . . . .659 

The United States Consulate ........ 660 

Women and Children waiting to be lowered down De Beers Mine . . 660 
Hoisting Double-decked Cage loaded with People and their Bedding . .661 
Boer Gun captured at Dronfield . . . . . . . .662 

Surrender of General Cronje . . . . . . . 663 

Lord Roberts and General Cronje ....... 664 

General Cronje as a Prisoner of War . . . . . . .665 

Reception of Lord Roberts and Lord Kitchener, Kimberley Town Hall . 666 

MAPS 

Blaeuw's Map of Africa ....... between 32-33 

Visscher's Map of Africa, 1662 ....... 33 

Africa from an Early Dutch Map . . . . . . .34 

Outline Copy of the Catalan Mappermonde, 1375 . . . 36, 37 

Outline Copy of the Map of Portolano Laurenziano, 1351 . . . 38 
Africa de Mappermonde, Juan de la Cosa, 1500 ..... 39 

Chart showing Method of Surveying Coast Lines ..... 46 



xviii ILLUSTRATIONS 

PAGE 

Map showing the Position of Ancient Ruins in Rhodesia .... 47 

Kimberley Mine between 276-277 

Diamond Mines owned by De Beers Consolidated Mines Limited between 316-3 I 7 
A Map to illustrate the Rhodesia Railway System . ... 592 

PHOTOGRAVURES 

FACING PAGE 

Portrait of Gardner F. Williams ....... Frontispiece 

Farmhouse on the Farm Constantia, near Cape Town .... 64 

La Rhone, Groot Drakenstein | 

Old Le Roux . . \' 

A View from the Kloof Road leading from the Upper Part of Cape Town . 86 

Zulu in War Attire ..... . .94 

The Author's Collection of Diamonds . . . . . . .154 

The Homestead of the Farm Vooruitzigt on which are De Beers and Kim- 
berley Mines . . . . . . . . .172 

Portrait of Sir Richard Southey . . . . . . . 1 80 

Natives seeking Work . . . . . . . . .188 

Kimberley Mine, 1872 I. 196 

Floating Reef, Kimberley Mine ........ 240 

Kimberley Mine, 1886 246 

Bultfontein Mine, 1878. . . . . . . .252 

Portrait of Cecil John Rhodes . . . . . . . .272 

Barnato's Residence, Kimberley ........ 296 

A Group of Officials 298 

Premier Mine, looking from Workings up through Incline where the Blue 

Ground is hauled ......... 344 

Premier Mine . . . . . .352 

Washing Machine with 1 8 Pans. Capacity, 6000 Loads, equal to 4800 

Tons in Ten Hours . . . . . . . . .374 

Waiting for News from the Mine, July 12, 1888 392 

Dutoitspan Road, Kimberley ......... 454 

The Gardens of the Intermediate Pumping Station of the Kimberley Water " 

Works Company ......... 466 

Piece of Blue Ground ......... 484 

A Thin Section of Diamond-bearing Rock (enlarged 3^ times) from the 

1320-Foot Level of the De Beers Mine ..... 486 

Cape Town ........... 562 

Mr. Rhodes's House 574 

Victoria Falls ........... 592 

The Terrace Garden 598 



THE DIAMOND MINES OF 
SOUTH AFRICA 

SOME ACCOUNT OF THEIR RISE 
AND DEVELOPMENT 




The Diamond Mines of South Africa 




CHAPTER I 

THE ANCIENT ADAMAS 

T the beginning of the last century, when the 
blinded Shah-Shuja sought refuge in the lair 
of the " Lion of Punjaub," Runjeet Singh, his 
chief treasure was the crystal pebble which Nadir 
Shah had snatched from the head of the last of 
the Great Moguls. 
For the sake of the pebble, Runjeet starved the wife and 
children of his friend until he was driven to lay the Koh-i-nur 
at the feet of his host. " At what 
price do you value it?" said the 
Lion, showing his teeth in a grim 
smile. 

"At good luck," replied the 
blind Shah, " for it has ever been 
the bosom companion of him who 
has triumphed over his enemies." 

It may have been the tradi- 
tional talisman of Carna, Rajah 
of Anga, fighting in legendary The Koh-i-nflr. (ow Cutting.) 

wars, hundreds of years before the great Achilles stormed and 
sulked under the walls of Troy. 1 From its earliest known 
appearance it had been so coveted that agas and sultans and 

1 "Tales from Indian History,"]. Talboys Wheeler, assistant secretary of 
the government of India in the Foreign Department, Calcutta, 1881 ; "The 
Great Diamonds of the World," Edwin W. Streeter. 




THE DIAMOND MINES OF SOUTH AFRICA 





rajahs and shahs had snatched it in the first spoils of victory, 
or tried to extort it by starvation or blinding or boiling oil or 
some other device of torture ; and the adventurous and blood- 
stained career of this famous diamond is only one of many like 

passages, for every precious 
stone of renown has a trail 
like a meteor. Some have 
gleamed weirdly in the eye- 
sockets of idols in Indian 
temples or flashed from 
the splendid thrones of 
emperors, or glittered in 
golden basins amid gems 
of every hue heaped up in 
tribute, or sparkled on the 
crests of warriors, the tur- 
bans of rajahs, the breasts 
.of begums, and the san- 
dals of courtesans. To 
win them temples have been profaned, palaces looted, thrones 
torn to fragments, princes tortured, women strangled, guests 
poisoned by their hosts, and slaves disembowelled. Some have 
fallen on battlefields, to be picked up by ignorant freebooters 
and sold for a few silver coins, and others have been cast into 
ditches by thieves or swallowed by guards, or sunk in ship- 
wrecks, or broken to powder in moments of frenzy. No strain 
of fancy in an Arabian tale has outstripped the marvels of fact 
in the diamond's history. 

Among all the stones that our world's fancy holds precious, 
the diamond stands preeminent. It is pure crystallized carbon. 
It crystallizes in almost all the forms of the isometric system, 
commonly the octahedral or dodecahedral, and frequently with 
curved faces. 1 Two pyramids with triangular sides and a 

1 Dana and others mention that diamonds in the form of cubes have been found. 
While one might expect to find a diamond in cubic form, as this is the fundamental 
form of the isometric system, still no specimen of this form has come under the 



i. A Black Diamond in Gold Setting. 2. Ordinary 
Window Glass. 3. A Pink Diamond. (Photo- 
graphed with the Roentgen Rays.) 



THE ANCIENT ADAMAS 3 

common base make up the octahedron. The dodecahedron has 
twelve rhombs or natural facets of lozenge shape. 

It is the most impenetrable of all known substances, for the 
edge of one of its facets will scratch the face of any other stone 
or the hardest steel. It is the most perfect reflector of light. 
It refracts entering rays more than any other translucent sub- 
stance except crocolite, the chromate of lead. 1 Chrysolite alone 
exceeds its dispersive power to dissolve white light into rainbow 
tints, but its combined powers of reflection, refraction, and dis- 
persion are unmatched. 2 Hence appears the play of color in its 
crystalline heart and the resplendent flashing of its radiant fire. 
It may be as purely transparent and colorless as a drop of dew, or 
it may display all the primary colors, such as red, orange, yellow, 
blue, and violet; so that, as John Mandeville quaintly observed, 
" It seems to take pleasure in assuming in turn the colors proper 
to other gems." 8 It is highly phosphorescent. Even the blackest 
of diamonds are transparent to the X-rays. No acid will mar it, 
no solvent will dissolve it. Its brilliance is undecaying, and ages 
might roll by without rubbing the minutest particle from its ada- 
mantine face. The diamond that gleamed with such strange fire 
in an idol's eye before the rising of the Star of Bethlehem may be 
sparkling to-day with more dazzling 
radiance in the crown of an emperor. 
Koh-i-nur and Darya-i-nur and Taj- 
e-mah and Regent and OrlofF and 
Sancy and Shah will shine no less re- 
splendent when the sovereigns that 
now treasure them shall be dust. The shah. 

observation of those whose duty it is to look over every stone that comes from the 
South African mines. The South African diamonds differ in appearance from those 
found in India or Brazil. They are bright and without any incrustation, and the 
imperfections, if any, are visible in their natural state. 

1 " Table of Indices of Refraction," Dufrenoy, p. 87. " Treatise on Gems," 
Feuchtwanger, New York, 1867. 

2 "Table of the Distinguishing Characteristics of Gems," Feuchtwanger, 
pp. 494-499. "Optical Properties of the Diamond," Sir David Brewster, 
Phil. Trans., VIII, 157, 1817. s " Le Grand Lapidaire," Paris, 1561. 




4 THE DIAMOND MINES OF SOUTH AFRICA 

" With the point of a diamond," Jeremiah (B.C. 600) says, 1 
records were graven when stones were writing-tablets ; but, 
unfortunately for our knowledge, the diamond did not tell its 
own story ; and it is, at best, a groping effort that would search 
out the rising of this gem through the mists of tradition. 

" Thou hast been in Eden, the garden of God ; every 
precious stone was thy covering, the sardius, topaz, and the 
diamond, the beryl, the onyx, and the jasper, the sapphire, 
the emerald, and the carbuncle. Thou wast upon the holy 
mountain of God ; thou hast walked up and down in the midst 
of the stones of fire." * 

How glowing are the words of the Prophet of the Captivity, 
declaring the vainglory forerunning the doom of Tyre's princes 

and people (588 B.C.). Did the three 
rivers of Eden flow through sands glit- 
tering with stones of fire ? Did the eating 
of a little green apple from the tree of 
knowledge open the eyes of the first 
woman of earth to the lure of the gems 
that are now so tempting to every daugh- 
ter of Eve ? If not, how long was it 

The Egyptian Pascha. before the topaz an( J the diamond, the 

emerald and the ruby and the sapphire were added to the fig- 
leaf covering of our first parents ? 

Multicycles of refining are needed for a clear perception of 
beauty. The aboriginal Adams and Eves did not have it. The 
children of the twentieth century will open their eyes to its 
light more quickly than those of the Stone Age, because the 
children of to-day inherit the quickened sense of unnumbered 
generations, and are taught to trace the range of beauty in 
nature and art. Prehistoric man, a weakling in perception, 
turned his eyes to the grand orb of the sun, rising above the 
horizon and flooding the earth with its rays, to the pale bow 

1 Jeremiah xvii. 

2 Ezekiel xviii. 13 and 14 (588 B.C.). Babylonian captivity of the Jews 
(588-537 B.C.). 




THE ANCIENT ADAMAS 5 

of the moon and the sparkling of the firmament of stars, to 
the ceaseless surge of the ocean and the mountain summits 
wreathed in clouds, to all the grander aspects and motions 
of nature, before his eyes were drawn to lesser things outside 
the petty circle of his rambling and the sating of his crude 
animal wants. Mayhap thousands of years of brutal life rolled 
by before the savage stooped to pick up any one of the gleam- 
ing pebbles which the fierce tiger spurned with bounding foot 
and the flying deer trampled heedlessly on the river's bank. 

Any one may guess, and any one's guess is as good as another's, 
what little pebble first drew the glance of the barbarian's eye or 
the stoop of the rover's knee. The first-known precious stones 
of the world were undoubtedly found on the face of the ground, 
without any wearisome digging or quarrying, as they lay shin- 
ing in the gravel, washed from hillsides over the plains, or along 
the courses of rivers swelled by floods and sweeping the par- 
ings of the earth's crust to the sea. Thousands of carnelians, 
garnets, jasper, amethysts, sapphires, rubies, and diamonds 
were picked up, maybe by children rummaging in gravel beds 
or the clefts of rocks, and thrown away as carelessly as splinters 
of flint, before one was preserved and prized. White and tinted 
shells were much easier to collect and pierce and link together, 
and rude armlets and leg-bands of copper and silver and gold 
were easily forged, and more to the savage taste than any neck- 
lace of stones. 1 

When some of the precious stones were lifted and borne 
away from their beds in drifts of gravel, they were valued first 
chiefly for the mystic powers attributed to pebbles of such rich 
hues, phenomenal hardness, and peculiar lustre. One of them 
would be worn in a pouch next to the bosom as an amulet or 
charm, averting peril, inspiring courage, healing diseases, repell- 
ing evil spirits, or winning the love of scornful maidens. Or, 
if any one of these magic stones was set to gleam in the buckle 
of a warrior's plume, it was less for a show of ornament than for 

1 "A Treatise on Diamonds and Precious Stones," John Mawe, London, 
1813. 



6 THE DIAMOND MINES OF SOUTH AFRICA 

its mystic shielding power and redoubling of valor. The tradi- 
tion of these virtues has passed from generation to generation, 
and still finds credence among the masses of Asia. The poor 
natives of India believe to this day in the efficacy of sapphires 
and rubies in purifying the blood, strengthening the body, 
quenching thirst, dispelling melancholy, averting danger, and 
assuring honor and fortune. The emerald in their eyes is 
potent to dispel bad dreams, give courage, and cure palsies, 
colds, and acute dysentery. The turquoise they say will brighten 
and heal weak and sore eyes, and serve as an antidote for veno- 
mous snake bites. 1 Like the other precious stones, the diamond 
was early endowed by fancy with medical virtues, and particu- 
larly prized as a safeguard from madness, in its power to " raze 
out the written troubles of the brain." 2 It was also believed to 
be potent to touch the heart, and there is a pretty conceit that 
the darts of Cupid were diamond tipped. Perhaps the passion 
of women for gems gave point to this fiction. 

As the diverse stones of fire became better known and more 
sharply distinguished, special significance was given to each by 
some nations of the East, associating them with the planets, 
the march of the seasons, or with various divinities. Sometimes 
they were of emblematic service. For the representation of the 
twelve tribes of Israel, twelve distinct gems were set in gold 
plates on the robe of the high-priest. 3 When the rise of letters 
and the fine arts brought the devising of symbols and graven 
inscriptions, the supposed potency of these stone amulets was 
increased by the craft of priests and sorcerers, cutting the face 
of the charms themselves or directing the hands of expert work- 

1 "Oriental Accounts of Precious Minerals," Journal of Asiatic Society of 
Bengal, August, 1832. 

2 "Treatise on Gems," Feuchtwanger, New York, 1867. 

8 Exodus xxviii. "Natural History of the Bible," Thaddeus M. Harris, 
Boston, 1820. "Precious Stones in their Scientific and Artistic Relations," A. 
H. Church, London, 1883. " De Duodecim Gemmis in Veste Aaronis." 
Epiphanius, 1565. John Peter Lange, Professor University of Bonn, in Schaff's 
' Critical, Doctrinal, and Homiletical Commentary ' ' on the Bible. 



THE ANCIENT ADAMAS 7 

men. The Chaldeans are especially charged with the fomenting 
of superstitions by the exaggeration of this conceit. These 
engraved stones served often as distinctive seals, and for con- 
venience in carrying and the gratification of a spreading taste for 
such ornaments, the talismans were set in rings and clasps. So 
Solomon's seal, summoning and mastering genii, was the wonder 
of legends, and so, too, the famous ring of Polycrates and the 
rival marvels of Oriental romancers familiar in the tales of the 
" Arabian Nights." 

As time and art disclosed more and more of the marvels of 
the stones of fire in the crust of the earth, the wonder grew and 
the supernatural potency of the various gems was more deeply 
impressed. Thus we reach the belief and tribute of the priest 
Onomacritus (500 B.C.), who declared of the lucent crystal, 
" Whoso goes into the temple with this in his hand may be 
sure of having his prayer granted, as the gods cannot withstand 
its power." Its use to concentrate the sun's rays as a burning 
glass was highly prized also in priestly ministrations. 

Onomacritus says crudely of this use that " when a trans- 
parent crystal is laid on wood, so that the sun's rays may shine 
upon it, there will soon be seen smoke, then fire, then a bright 
flame." Fire kindled through this agency was holy in the 
sight of priests and people, and no burnt offering was so pleas- 
ing to the gods as one set in these sacred flames. 

The precious stones are so greatly dependent upon the ad- 
vance in the art of polishing and cutting for the revelation of 
their qualities and beauty that it was doubtless long after their dis- 
covery before they came into any considerable use as ornaments. 
Their hardness defied, at first, any effort to fashion their shape 
with primitive tools. The most that could be effected was the 
rude polish that might be obtained by the tedious rubbing of 
the face of one stone against another. But, as time went on, 
the lines of natural cleavage were noted, and grinding wheels 
in the hands of skilful artisans gave a smooth face to the natural 
contours of the softer stones, and, later, even to the sapphire 
1 " Precious Stones and Gems," Streeter. 




8 THE DIAMOND MINES OF SOUTH AFRICA 

and diamond. With the advance in art the demand for precious 
stones increased apace, and, to meet the demand, keener and 
wider ranging searches developed new and greater supplies. 

There is a certain tracing of the use of 
precious stones for ornaments to the ancient 
Babylonian civilization, whose existing ruins 
extend back to from 6000 to 7000 years B.C. 1 
Babylonian lapidaries were cutting and polish- 
ing carnelians, sards, onyx, and rock crystals 
before the Egyptians had advanced beyond 

The Polar Star. i c \ r T-I i 

the carving or their sort steatite. 1 hen the 
Phoenicians drew from all parts of the known earth its treasures. 2 
So Ezekiel testifies of Tyre : " Syria was thy merchant by 
reason of the multitude of wares of thy making : they occupied 
in thy fairs with emeralds, purple and broidered work and fine 
linen and coral and agate. The merchants of Sheba and Raamah, 
they were thy merchants : they occupied in thy fairs with chief 
of all spices, and with all precious stones, and gold." 

Judea had some share of this stream. The Queen of Sheba 
bore a "great store of precious stones " to Solomon (B.C. 101 5-975) 
with her tribute of gold, 4 but this was a 
trivial trickle compared with the flow to 
Phoenicia and Babylonia. Long before the 
days of the Captivity (B.C. 598),' the robes 
of the princes and nobles of these rich realms 
were glittering with jewels, and their gor- 
geous array was the marvel of the poor The Hope Blue, 
exiles, crying with the voice of their prophet, Ezekiel : " Every 

1 " Encyclopedia of Religious Knowledge," SchafF-Herzog. "Archaeology 
of the Past Century," Professor W. M. F. Petrie. 

2 The Story of the Nations, " Phoenicia," George Rawlinson, M.A. 

3 Ezekiel xxvii. 22. 

4 " Old Testament History," William Smith. "Precious Stones in the 
Scriptures," R. Hindmarsh, London, 1851. 

5 Date of removal of Jehoiachin, according to Prideaux and to Clinton. Ewald 
makes the date 597 B.C. 




THE ANCIENT ADAMAS 9 

precious stone was thy covering. Thou hast walked up and down 
in the midst of the stones of fire." As tradition placed the garden 
of Eden in the valley of the Euphrates, Ezekiel makes the gar- 
den typical of the splendor of Babylon in his fervid outpouring. 

How the stones of fire were brought into being in the garden 
of Eden or elsewhere, Ezekiel was not moved to reveal, and the 
savants that have sought to tell are but groping seers. When a 
sprinkling of stones was uncovered by the rains and floods, or 
dug and washed from the beds of gravel, or traced by rude min- 
ing through clay or conglomerate layers or enclosing rocks, there 
was still no widespread knowledge of the deposits, and even 
among the most familiar with the search there was ever the 
hope of finding, some day, some marvellous store. Hence 
sprung up the romances. Even in the days when the sharp 
tooth of history had cut into legends, a story was told of the 
climbing of Zulmat by the great Alexander, to the rim of the 
inaccessible valley, where, beneath sheer precipices, glittered a 
coverlet of the stones of fire. There was no way of winning the 
diamonds that glowed so temptingly except by flinging down 
masses of flesh and waiting for swooping vultures to bear the 
lumps up to their perches on the mountain with precious stones 
sticking in the meat. 1 

Sindbad the sailor had this tale in mind fortunately in his 
second voyage. It will be remembered that he was stranded 
by shipwreck on a desert island and carried away by the flight 
of a gigantic rukh to the top of a distant mountain. From 
this mountain he descended into a neighboring " valley, exceed- 
ing great and wide and deep and bounded by vast mountains 
that spired high in air." Walking along the wady, he found 
that " its soil was of diamond, the stone wherewith they pierce 
minerals and precious stones and porcelain and the onyx, for 
that it is a dense stone and a stubborn, whereon neither iron or 
hardhead hath effect, neither can we cut off" aught therefrom, 
nor break it save by means of lead stone." 

1 " Oriental Accounts of Precious Minerals," Journal of Asiatic Society of Ben- 
gal, August, 1832. 



io THE DIAMOND MINES OF SOUTH AFRICA 

Luckily for the sailor, his descent was by day, for " the val- 
ley swarmed with snakes and vipers, each as big as a palm tree, 
that would have made but one gulp of an elephant ; and they 
came out by night, hiding during the day, lest the rukhs and 
eagles pounce on them and tear them to pieces." In view of 
the horrid prospect of soon dropping through the throat of one 
of these snakes, Sindbad began to wish that he had not flown 
away from the island, where he was, at least, out of reach of vast 
vipers, but he soon bethought himself of the old story of the 
valley from which diamond-studded meat was "plucked by 
eagles." So he quickly filled his pockets and shawl girdle and 
turban with the choicest diamonds. Then he put a piece of raw 
meat on his breast and lay down on his back. Soon a big 
eagle swooped into the valley, clutched the meat in his talons, 
and flew up to a mountain above, " where, dropping the carcass, 
he fell to rending it," leaving the lucky sailor to scramble off" 
with his booty. He gave a parcel of the diamonds to the dis- 
appointed merchant, who had cast down the meat, but he had 
stuffed his clothes so full of the gems that he went home, after 
some strange sight-seeing, with a great store of diamonds and 
money and goods. 1 

This amazing tale is less teeming with interest than it was in 
the days when it was first told, for, even hundreds of years 
afterwards, diamond-lined valleys and monstrous rukhs and 
snakes that could gulp down elephants were not beyond cre- 
dence. If in valleys there might be a diamond lining, why 
should there not be a massing of diamonds and rubies in the 
dwellings of genii in caves, awaiting the entry of some lucky 
Aladdin ? Oriental fancy, teeming with visions, disdained any 
curbing within the petty confines of crawling experience, and was 
prolific in marvels far more pleasing to the masses that egged 
on the story-tellers with craving credulity. Who then could 
explode these bubbles with any sharp prick of positive contra- 
diction ? Even if in all known fields the precious stones were 
gathered by toilsome searches only rarely rewarded, who had the 
1 "Arabian Nights," Lady Burton's edition, Vol. Ill, pp. 476-482. 



THE ANCIENT ADAMAS n 

range of knowledge to deny the possible existence of caverns 
filled with rubies or mountain summits studded with diamonds ? 

Seeing that to this day so little can be asserted positively of 
the forming of the precious stones scattered in the earth's crust, 
it is not surprising that the origin of the stones of fire has been, 
from the first, a baffling puzzle and a fountain-head of conflict- 
ing surmises. Some wondering people viewed them as splin- 
ters dropping from the stars, and some, as the creations or 
transformations of genii. Some Hindoo miners still believe 
that diamonds grow like onions, though much less quickly, and 
that their age is marked by the difference in their size and 
quality. Others suppose the common rock crystals to be 
immature diamonds, and the distinction is marked by calling 
the rock crystal kacha (unripe), while the diamond is pakka 
(ripe). 1 

For the ripening of the crystals and the quickening of their 
seeming inward fire, the lightning bolts, that sometimes rived 
the ground, were thought to be potent. Others again, observ- 
ing the liquid purity and likeness which is marked to this day 
in the term " diamonds of the purest water," attributed the 
forming of the crystals to the supernormal trickle and hardening 
of dewdrops. It is of this fancy that Dryden makes poetic use 
in his likening of the tears of Almahide : 

" What precious drops are those, 

Which silently each other's track pursue, 

Bright as young diamonds in their infant dew ? ' ' 

Bizarre speculation was stretched even to the point of attrib- 
uting to these strange crystals animal instincts and reproductive 
powers. Thus Barreto is quoted in the dictionary of Antonio 
de Moraes Silva as saying : 

" Que os diamantes se unem, amam e procream." 3 

1 "Oriental Accounts of Precious Minerals." Translation by Rajah Kalikis- 
ken, Asiatic Society of Bengal. 

2 "The Conquest of Granada," Second Part, Act III, Scene I, Dryden. 

3 " Commonplace Book," Second Series, p. 668, Southey. 



12 THE DIAMOND MINES OF SOUTH AFRICA 

The tradition of the generative power of this marvellous 
crystal originates with the Hindoos, and to this day the 
natives of Pharrah will affirm that the diamond beds yield 
fresh supplies of well-grown stones at intervals of from fifteen 
to twenty years. 

It is seemingly hopeless to attempt to fix with any certainty 
the time when the diamond was first singled out from the peb- 
bles in which it lay, and was prized by any one, or even when it 
entered the list of gems known to the chief nations of Asia. 
Traditions coming down through the mists of legendary ages are 
conflicting and uncertain reliances at best. The ancient writers 
add to this perplexity by loose or erroneous descriptions when 
the advance of the science had not marked precise distinctions 
of structure and composition. Thus the Carbunculus of Pliny 
was probably stretched to cover the spinal or Balas ruby, the 
garnet and other red stones, besides embracing the Anthrax of 
Theophrastus or our modern ruby. Many ancient writers con- 
founded also under the general term Smaragdus various dis- 
tinct minerals of green color, not only the true emerald, but 
green jasper, malachite, chryscolla, and fluor spar. 1 Among the 
common people, pretending to no mineralogical knowledge, 
there was less thought of distinction, and, in days approaching 
our own, Tavernier observes in his travels, A.D. 1669, after 
describing the true ruby of Pegu, in Ceylon, " the fatherland 
of rubies," that " all other stones in this country are called by 
the name Ruby, and are only distinguished by color, thus, in 
the language of Pegu, the sapphire is a Blue Ruby," etc. 2 This 
confusion is not surprising, and a much more discreditable one 
occurred within the last thirty years in the sensational touting 
of the discovery of rubies in the garnets of the" Macdonnell 
Ranges in South Australia. It seems highly probable that the 
stone of exquisite blue, now particularly distinguished as the 
typical sapphire, was the ancient Hyacinthus ; and the Sap- 
phirus of the ancients certainly included the lapis lazuli and 

1 " Precious Stones and Gems," Streeter, London, 1892. 

* "Voyages en Turquie, en Perse et aux Indes," Paris, 1676. 



THE ANCIENT ADAMAS 13 

covered the range of corundums of every tint except red. Thus 
green sapphires are noted, although very rarely, and yellow and 
gray, as well as pure white or colorless, and this stone is pre- 
sumed by Streeter and other investigators to have been the 
" adamas " first known to the Greeks. 1 

There can be no question that sapphires or corundums of 
varied hue were much more common than diamonds in the hands 
of the merchants of the East or any other ancient collectors before 
the Christian era. The sapphire was, indeed, one of the most 
widely known of all gems, and how highly it was valued may be 
surmised from the dignity given to it by the sacred writers. The 
prophet Ezekiel likens to a " Sapphire stone " the appearance of 
the throne in the firmament above the cherubim. Job makes 
it the representative of all gems in his splendid description of the 
daring of miners. 2 

Like the sapphire, the diamond is repeatedly referred to by 
the Hebrew writers. It formed one of the typical stones in the 
high priest's breastplate, and Ezekiel puts it in the first rank of 
the stones of fire. Jeremiah speaks of the sin of Judah as written 
with the point of a diamond, " puncto adamantinis" of the Latin 
Bible, but Streeter holds that this pen point was probably a 
corundum and not the true diamond. 3 

This is a stretch of assumption largely based upon the lack 
of any precise description applying to the diamond until close to 
the beginning of the first century of our reckoning. Adamas, 
the indomitable, the adamant of the ancients, was the name given 
to the diamond because of its distinguishing hardness. Pliny 
was greatly impressed by what he heard of this characteristic, 
but obviously knew little or nothing of the stone by personal 
handling or test. For he wrote down soberly : " The most 
valuable thing on earth is the Diamond, known only to kings, 
and to them imperfectly. It is only engendered in the finest 
gold. Six different kinds are known, among these the Indian 

^'Traite de Mineralogie, avec application aux Arts," Brongniart, Paris, 
1807. 

2 Job xxviii. i-i I. 3 " Precious Stones and Gems," Streeter. 



14 THE DIAMOND MINES OF SOUTH AFRICA 

and Arabian of such indomitable, unspeakable hardness, that 
when laid on the anvil it gives the blow back in such force as to 
shiver the hammer and anvil to pieces. 1 

Unfortunately for the aim of identifying the diamond with 
the references to the ancient adamas, the term was commonly and 
loosely applied to any substance of peculiar hardness. So moun- 
tains of iron-stone, like unto that upon which the ship of Sindbad 
was dashed, were called adamant, and so too were the arms and 
armor of gods and heroes. Addison only transmits a tradition 
in the fine lines of his poem 

" And mighty Mars, for war renowned, 
In adamantine armor frowned." 2 

In Homer, as Streeter notes, adamas occurs only as a per- 
sonal name, and in Hesiod, Pindar, and other early Greek poets 
it is used to signify any hard weapon or metal like steel or an 
alloy of the harder metals. 8 No distinct identification of the 
diamond with adamas appears, according to Streeter's view, until 
the first century A.D., in the writings of the Latin poet and astron- 
omer Manilius, and his contemporary Pliny (A.D. 62-114). In 
the fourth book of Manilius's poem " Astronomicum," occurs 
this line, " Sic Adamas, punctum lapidis, pretiosior auro," which, 
Streeter says, " is supposed to be the earliest indubitable reference 
to the true diamond." It is difficult to see how this " stone's 
point, more precious than gold," is any more distinct and indubi- 
table in its reference to the diamond than the diamond pen point 
of Jeremiah hundreds of years before. But Pliny, with all his 
erroneous amplifications, unquestionably describes the true Indian 
diamond as " colorless, transparent, with polished facets and six 
angles ending either in a pyramid with a sharp point or with two 
points like whipping tops joined at the base." * 

1 "Historia Naturalis," XXXVII, 15. 

2 Poem addressed to Sir Godfrey Kneller, referring to William III. of England. 

3 " aSa/jas ytvos otSvrjov," -^schylus. See Stanley's Commentary on ^Eschy- 
lus, "Prometheus Vinctus." 

4 Plinii Secundi (Caii), " Naturalis Historia," XXXVII, 15. 



THE ANCIENT ADAMAS 15 

In view of the hardness of the sapphire, so great that it will 
scratch every other precious stone except the diamond, it is there- 
fore contended that this was the stone known to the earliest 
Greek writers as adamas. 1 This may be so, and it cannot be 
doubted that, even at a much later day, a white corundum or a 
pale yellow topaz or a good rock crystal often passed for a dia- 
mond in the hands of collectors or in the sharp practice of gem 
selling. Whatever may have been the blundering of the Greeks 
or the application of adamas, there is, nevertheless, no sufficient 
reason in this for questioning the probability that genuine dia- 
monds were found in the gravels of India many centuries before the 
Christian era. As far back as tradition goes the largest stones 
were particularly prized by the native princes, 
and were strictly exacted in tribute from the 
diamond-bed washers. But the smaller stones 
were less jealously guarded, and may readily 
have found their way into the hands of traders 
with the other peoples of Asia or with Egypt. 
It seems most probable that the Jews derived 
their first knowledge of precious stones from The Empress Eugenie - 
the Egyptians chiefly, for the Hebrew names of the stones are 
of Egyptian derivation. 2 Thus there is no approach to certainty 
for the assumption that the stones called diamonds in the English 
version of the Hebrew Scriptures were not rightly named, or 
that allusions to the diamond in other ancient writings were 
wholly unreliable or mistaken. 

The main support for the questioning of the mingling of 
diamonds with the other gems noted by the ancient writers is 
the apparent failure to uncover diamonds in the excavations on 
the site of ancient temples and cities where other precious stones 
are brought to light. Thus emeralds and other gems in various 
settings have been exhumed from the volcanic overflow that 

1 "History of Stones," Theophrastus. Edited by Sir John Hill, London, 
1746. " Elem. de Min.," Lessing, II, 61. "The Great Diamonds of the 
World," Streeter. 

2 " Encyclopedia of Religious Knowledge," Schaff-Herzog. 





i6 THE DIAMOND MINES OF SOUTH AFRICA 

buried Herculaneum and Pompeii, from the ruins of old Rome, 
and the tombs of Egypt. 1 In the course of explorations on the 
site of Curium and other ancient towns in Cyprus, scarabs 
and scaraboids of agate, onyx, jasper, and variously tinted car- 
nelians were found, as well as gold ornaments, relics traced to the 
days of Eteandros, king of Paphos in the seventh century B.C. ; 
but no diamonds were unearthed in this collection.' 2 Nor is 

there record, as yet, of the discovery of 
diamonds in the explorations in Baby- 
lonia. 3 

But this is, at most, evidence pointing 
to what is undoubted, the comparative 
rarity of the diamond among the gems 
that served as amulets or ornaments for 
the people of western Asia, northern 

The Nassak. n r T> 

Atrica, or southern Europe prior to the 

Christian era and for centuries afterward. Pliny expressly asserts 
this rarity in his allusion to the diamond ; but the fact that the 
gem was scarce, outside of India, is entirely compatible with its 
occasional inclusion in the collections of sovereigns, which the 
same writer remarks, and the high value set upon it would 
naturally limit its use as an ornament. 

It is impossible to mark with any precision in what district 
of India a search for diamonds first began. Rajah Sourindo 
Mohun Tagore, in his account of the precious stones of India, 
gives the names of eight localities in which diamonds have been 
found according to tradition or more certain report. These are 
Harma (Himalayas), Matanga (Kistna), and Godaveri (or Gol- 
conda), Saurashtra (Surat), Paunda (probably including the Chutia 
Nagpur Province), Kalinga (the tract between Orissa and the 
Godaveri), Kosala (the modern Ajodhya or Berar), Vera Ganga 

1 Clarke's Travels," Vol. VIII, p. 150. 

2 Story of the Nations, " Phoenicia," George Rawlinson. 

3 "Nineveh and Babylon," pp. 160161, 602 et seq., Layard ; " Arch- 
asologyofthe Past Century," Professor W. M. F. Petrie. 



THE ANCIENT ADAMAS 



(the Wemganga), and Saubira (the stretch between the Sarhund 
and Indus rivers). 1 

According to this showing the diamond is scattered over a 
wide ranging region, but it occurs everywhere in one of two 
comprehensive formations, alluvial or otherwise disintegrated 
surface deposits, and conglomerate rocks of far receding geo- 
logic antiquity, belonging to the Vindyhan formation, which 
borrows its name from the Vindyhan Hills of old geographers. 2 
It seems reasonable to presume that the surface wash comes 
from the disintegration of the seat of the diamond in con- 
glomerate beds, for even in alluvial gravels there are fre- 
quently no diamonds found outside of a conglomerate of 
rounded pebbles and sandstone breccia. It is likely that the 
first diamonds were taken from the surface wash and that the 
more solid breccia was opened later. 

In some of the diamond-bearing districts of India to-day 
the native villagers are searching for diamonds exactly as their 
fathers did in days of remotest tradition. After a heavy rain 
that washes away loose soil, a 
sprinkling of diamonds may be 
found in exposed sandstone brec- 
cia, and sharp-eyed Hindoos 
scrape the face of the ground for 
the precious crystals. 

Along the banks of the Kistna 
and Godaveri rivers the Golconda 
of tradition outstretched, and this 
diamond-studded ground came 
later into the hands of the Nizam 
of Hyderabad, and was included The Great Mogul. 

in the bounds of the Madras Presidency. Here, it is claimed, 
was the bed of the Koh-i-nur and Regent and Great Mogul, 
and others of the jewels most renowned in history and romance. 
Here, of a certainty, was the richest diamond field in India, in 

1 "Mani Mala," Calcutta, 1879. 

2 "Manual of Geological Survey of India," Professor V. Ball, Vol. III. 




i8 THE DIAMOND MINES OF SOUTH AFRICA 

the days of Tavernier's travels (1669 A.D.). Here was the 
famous mine, " Gani-Coulour," that he saw, where sixty thou- 
sand natives were then at work, and " Gani-Parteal," and 
twenty more of lesser note. 1 Gani-Coulour has probably been 
identified with the modern Kolur on the Kistna, Gani being 
simply a slight change of the Persian " Kan-i " or "mine of," 
so that Gani-Coulour is the mine of Kolur as Gani-Parteal is 
the mine of Parteal. 2 The surface ground of this district along 
the rivers is a black " cotton soil " washed down by floods, and 
underlying this at an average depth of twenty feet is a layer of 
broken sandstone, quartz, jasper, flint, and granite, interspersed 
with masses of calcareous conglomerate, forming the stratum in 
which the diamonds were embedded. When the black soil had 
been dug up laboriously and carried away, the diamond-bearing 
layer was exposed, and was removed, piecemeal, to level stretches 
of ground or prepared floors, where it was scraped and picked 
over by hand to find the diamonds. 

The whole of this rich mining district and a tract stretching 
for many miles away was loosely called Golconda, or the King- 
dom of Golconda, by foreign traders and travellers, because the 
town of Golconda was its capital and the trading centre where 
the diamonds from the mines were chiefly bought and sold. 
The only mark of this old mart to-day is a deserted fort near 
Hyderabad, but its fame will endure until traditionary Golconda 
ceases to be a standard of riches. 

Next in importance and prestige to the mines of Golconda 
was the diamond field of Sumbulpur, in the Central Provinces, 
between the rivers Mahanadi and Brahmini. The diamonds of 
this district were remarkable for their purity and beauty, though 
no very large crystals have been traced to this region, and the 
few which the washings still yield rank with the finest of the 
Indian stones. Here the precious stones were found chiefly 
along the course of the Mahanadi, in a stratum of tough clay 
and pebbles stained reddish by iron oxide. At the opening of 

1 "Voyages en Turquie, en Perse et aux Indes," Tavernier, Paris, 1676. 

2 " Manual of Geological Survey of India," Vol. III. 



THE ANCIENT ADAMAS 19 

the dry season, thousands of villagers, men, women, and chil- 
dren, began to search every cleft and cranny in the river beds 
for diamonds. With ankovas, or light picks, the men broke 
and scraped out the diamond-bearing bed and piled the broken 
ground on the river bank. Then the women scooped up 
ground from the heaps with their daers. These were shovel- 
shaped boards, about five feet long, with ridged sides and hol- 
lowed in the centre. Resting one end of the daer on the 
ground and tilting the other slightly, they washed away the clay 
and sand and picked off the rock splinters and larger pebbles. 
After this rude sorting they spread out the finer gravel on a 
smaller board, the kootla, and scraped it over very carefully to 
separate the diamond crystals and grains of gold. When there 
was a level stretch along a bank, the native workers would some- 
times make an enclosure on this flat, with a low wall pierced at 
several points by small waterways. Then they would dump 
the diamond-bearing ground into this shallow basin and wash 
away the clay and dirt with running water. After two or three 
washings they would pick out the larger stones from the cleaned 
gravel, and dry the remainder, to be picked over on their kootlas 
or on any smooth, hard flooring. 

Perhaps the most laborious diamond digging in India has 
been in the pits of Panna and neighboring villages in the Prov- 
ince of Bundelkhund. Here the diamond-bearing conglomerate 
was buried under a cover of heavy ground, ranging in places 
over thirty feet in thickness. To reach the diamond strata large 
pits were dug, with inclines leading to the bottom in or below the 
conglomerate. There was no drainage, and the diamond diggers 
were forced to work in the rainy season knee-deep in water, 
breaking the conglomerate, and filling baskets which were hauled 
by hand to the top of the pits. In this primitive fashion the 
diamond beds of India were opened, and diamonds are to-day 
won by these simple methods or others essentially similar. 1 

1 "A Treatise on Diamonds and Precious Stones," John Mawe, London, 
1813. " A Treatise on Gems," Feuchtwanger, New York, 1867. "Precious 
Stones and Gems," Streeter, London, 1892. 



20 THE DIAMOND MINES OF SOUTH AFRICA 

Color and size were the chief distinction in diamonds, as in 
the other precious stones, in the early days before the advance 
of the art of diamond cutting which has added so greatly to the 
brilliancy and beauty of this gem. Centuries ran by before 
the ancient lapidaries attempted more than the polishing of the 
surfaces of the natural facets of the crystal, though the compara- 
tive ease with which this hardest of stones may be split by fol- 
lowing the natural cleavage lines may have been observed. Size 
was rated so highly by the Hindoos in valuing a gem that the 
conception of increasing the worth of a jewel by cutting away 
the greater part of it would not have been tolerated even if it 
had been feasible. When cutting to a limited extent began to 
be practised in India, it was generally unsymmetrical and unsci- 
entific, as the oldest known diamonds bear witness, and there 
was comparatively little advance for many centuries, as every 
celebrated gem of Indian workmanship plainly shows. 1 But 
even with imperfect cutting and crude polishing the inherent 
beauties of the ancient stones were more or less fully disclosed. 

In the mines of Panna there were four noted divisions in 
grading. Clear and brilliant stones were in the class Motichul, 
Mansk was the class name applied to diamonds of greenish tint, 
Panna to light yellow, and Bunsput to sepia colored stones. 2 
In India at large there was a comprehensive divisional grading 
corresponding to the main caste distinctions, the "twice-born," 
priests, warriors, and merchants, and the " once-born," tillers of 
the land. 3 The Brahmans were the diamonds of highest range, 
clear and colorless crystals ; the Kshatriyas, clear crystals, amber 
tinted or of the color of honey ; Vaisyas, the cream colored ; and 
the servile Sudras, the grayish white stones. Grades in rank 
were more minutely marked in the rubies of the famous Badak- 
shan mines in Persia, where the common people believed that 
the precious stones were deposited in the " rag-i-lal " or parent 
vein in successive layers. The outside layer contained the small 

1 "A Treatise on Diamonds and Pearls," David Jeffries, London, 1751. 

2 " Precious Stones and Gems," Streeter. 

3 "Annals of India," J. Talboys Wheeler, Calcutta, 1881. 



THE ANCIENT ADAMAS 21 

and imperfect stones styled piadehs, foot soldiers ; the next, a 
better class of stones called sawars, horse soldiers; and so on 
through layers of amirs, bakshis, and vazirs until a single stone 
was reached, transcending all in size and beauty, which the min- 
ers polished dutifully, and took in tribute to their sovereign. 1 

With the expansion of Greek commerce and the entry of 
Greek mercenaries into the employ of satraps in Asia Minor 
(about 500 B.C.), the riches of the Orient were made known, and 
precious stones began to pass into Europe. Herodotus, 484 B.C., 
was first of the early Greek writers 2 to mark particularly the dis- 
plays of precious stones in palaces and temples the signet rings 
of Darius, the magnificent emerald in the ring of Polycrates, and 
the marvellous show of the emerald column in the temple of 
Hercules in Tyre, gleaming like a pillar of green fire at night. 
This fiery column has a certain likeness to the traditional stone 
as big as an ostrich egg, to which homage was paid as the " God- 
dess of Emeralds" by the people of the Manca Valley in Peru. 
Sceptics would clip the marvel of both by substitution of beryl, 
or aquamarine, or colored glass ; but this trimming of legend 
does not question the extraction of true emeralds from mines in 
Upper Egypt, or the superb yield of the deposits in Peru and 
New Grenada. 8 

The conquests of Alexander the Great (334-323 B.C.) made 
the Greeks familiar with the precious stones of India as well as 
of Western and Central Asia. His successors revelled in pro- 
fuse displays of jewelled rings and bracelets, and wine cups and 
candelabra, in luxurious banquets. Pliny tells a glowing tale of 
a statue of Arsinoe, wife of Ptolemy Philadelphus (283 B.C.), 
four cubits in height, made of topazon. 4 The true topaz was 
undoubtedly known to the ancient Egyptians, and is still obtained 
at Risk Allah near the old emerald mines of Jebel Zabara ; but 
the Oriental topaz is presumed to have been the yellow sapphire ; 

1 "Oriental Accounts of Precious Minerals," Journal of Asiatic Society of 
Bengal, August, 1832. 2 Rawlinson's "Herodotus." 

3 Brun's "Travels." Rawlinson's "Herodotus," II. 44. Prescott's "Con- 
quest of Mexico." 4 "Historia Naturalis," XXXVII, 32. 



22 THE DIAMOND MINES OF SOUTH AFRICA 

and the Greek topazios, the yellowish green chrysolite or the 
peridot, of deeper green tint. The word is derived from T07raa, 
" to seek," because the traditional source was an island in the 
Red Sea, often difficult to reach through its envelope of fog. 1 
The loose use of the term by Pliny and other old writers makes 
it impracticable to mark with any certainty from what greenish 
hued stone Arsinoe's statue was cut. Still, in spite of current 
exaggeration and confusion of distinctions, there can be no doubt 
of the rising production and circulation of the precious stones. 

With the spread of the Roman Empire prodigality in dis- 
plays ran riot. After Pompey's victory over Mithradates, 
(B.C. 66) precious stones and pearls poured into Rome and the 
demand of vanity rose to a passion. 2 Pliny writes : " We drink 
out of a mass of gems crusting our wine bowls, and our drinking 
cups are emeralds." To heighten the wonder he tells in his gos- 
siping way how emeralds were set as the eyes of a lion sculptured 
in marble on the tomb of King Hermias in the island of Cyprus. 
So great was the size and so piercing the light of these emerald 
eyes that the tunny fish in the surrounding sea were frightened 
away until the fishermen of Cyprus put common stones in place of 
the dazzling gems. Later scepticism would make these emerald 
eyes of malachite, for copper ores were of common occurrence in 
Cyprus 8 and the glory of the emerald was scattered by loose usage 
over green fluor spar, jasper, aquamarine, malachite, and perhaps 
even green glass. There is also a shaking of the marvel of the cups, 
holding a pint, that were made out of solid carbuncles ; for these 
are supposed to be cuttings from the common garnets of the Bar- 
bary coast, flowing out from Carthage in such profusion that the 
carbuncle was called " the Carthaginian stone." 4 

Beryl was largely used in the ornamentation of cups and 

1 Diodorus Siculus, Lib. Ill, c. 38. Jameson's " Mineralogy," p. 48. 
Kidd's " Mineralogy," I, 121. 2 << Historia Naturalis," XXXVII, 6-7. 

' Cleaveland's "Mineralogy," p. 565. Theophrastus, " De Lapid.," 
c. 49. 

4 "The Story of Carthage," p. 121, Alfred J. Church, M.A. "Story of 
the Nations." 



THE ANCIENT ADAMAS 23 

for cameos ; l and carnelian was particularly prized as a base for 
the engraving of seals or cameos, sometimes elaborately pictorial. 
The great scarab in the Prussian cabinet, representing the five 
heroes of Thebes, is a recognized masterpiece of old Etruscan 
art, and a deep-cut carnelian once belonging to Michael Angelo 
portrays the birthday festival of Dionysius. 2 Amethyst ranked 
with carnelian as a favorite stone with engravers, and it was of 
peculiar traditional service in the fashioning of drinking cups, 
from its supposed checking of drunkenness, whence its Greek 
name, a, " not," and fj.tdva), " to intoxicate." Opals were 
placed in the first rank of gems, and Pliny tells of a senator, 
Nonius, who bore banishment and the loss of all his estate 
rather than the sacrifice of his opal ring to the greed of Mark 
Antony. 8 

Pearls were even more highly valued and lavishly displayed 
than any of the precious stones. Swelling the yield of the 
Mediterranean shores there flowed into Rome a profusion of 
still finer pearls from the Persian Gulf and Ceylon, to be set in 
necklaces, bracelets, earrings, and clasps of all kinds. Rich robes 
were bespangled with jewels, and it is reported that Lollia Pau- 
lina, the wife of Caligula (A.D. 37-41), wore a dress covered with 
pearls and emeralds. Cleopatra's famous pearls were said to 
have cost her _8o,ooo. Julius Caesar (B.C. 102-44) gave Servilia, 
Cato's sister, a pearl valued at over ^50,000, and Nero dropped 
handfuls of pearls in the laps of his mistresses (A.D. 54-68). 

From personal adornments, the decoration of arms and trap- 
pings, and the embellishing of banquets, the use of gems spread 
to the mounting of pictures in frames studded with precious 
stones, and the ornamentation of statuary. Nero viewed the 
combats of gladiators in a mirror of jewels, 4 and Constantine 

1 " Historia Naturalis," XXXVII, 20. 

2 "A Treatise on Gems," Feuchtwanger. 

3 Historia Naturalis," XXXVII, 21. 

4 Ibid. XXXVII, 1 6. Beckmann thinks the mirror of Smaragdus in which 
Nero gazed may have been green obsidian, green jasper, or even green glass. 
"History of Inventions," III, 177. 



24 THE DIAMOND MINES OF SOUTH AFRICA 

challenged the splendor of Oriental monarchs by his entry into 
Rome in a chariot of gold sparkling with precious stones 
(A.O. 310-337). 

Amid all this profusion, in which millions of sesterces were 
lavished, the diamond is noted only by rare allusions. This is 
probably accounted for by the check in the advance of lapidary 
art on reaching a stone of such indomitable hardness. Even 
the diamonds set in the clasp of the regal mantle of Charlemagne, 
after the opening of the ninth century, show only a partial polish- 
ing of the natural planes of the crystals. There was no scientific 
cutting of facets to heighten the brilliancy of the stone until the 
fifteenth century. When artificial shaping was attempted before 
that time, it did not go beyond the production of a flat top or 
table, or a convex surface, with a truncated pyramid as a base. 
Even when a large number of facets were cut, as was sometimes 
done by East Indian lapidaries, there was no scientific propor- 
tioning, as was signally shown in the instance of the remarkable 
stone known as the " Beau Sancy," which came into the posses- 
sion of Charles the Bold, Duke of Burgundy. It was the recut- 
ting of this stone in 1465, by the true artist Louis de Berquem 
of Bruges, that marks the rising of the modern art that has 
enhanced so immensely the resplendence and beauty of the 
diamond, and established its place securely as the chief among 
gems that are prized for adornment. 

Then begins the entry of the famous diamonds passing over 
the face of Europe with meteoric trains of adventure. The 
Beau Sancy glitters for a moment in the splendid array led by 
Charles the Bold against the Swiss peasants. On the bloody 
field of Granson (3d March, A.D. 1476) where the best knights 
of Burgundy were killed or put to flight by the mountaineers, 
the jewel that might ransom a king is trampled under foot in 
the rout. A Swiss soldier picks it up. It is no more in his 
eye than a bit of glass which he is well pleased to sell for a florin 
to a priest. Philip de Commines says that the priest knew no 
more of its value than the soldier, and thought he did well to 
make a franc by selling the diamond to the burghers of Berne. 




THE ANCIENT ADAMAS 25 

There the diamond disappears. One current story makes it 
reappear one hundred years later in the possession of the king 
of Portugal, who pledges it with other jewels for a loan from 
Nicholas Harlai, Seigneur de Sancy, and treasurer of the king 
of France. M. de Sancy soon buys it outright for one hundred 
thousand francs and loans it to sparkle for a time on the head 
of his king, Henry III. (A.D. 1574-1589). 
When Henry of Navarre comes to the throne 
(1589), M. de Sancy sends the diamond to 
him by a trusted servant. Thieves waylay 
and kill the messenger, but the precious stone 
is seemingly not in his keeping. So his body 
is thrown into a grave hastily made by his 
murderers. When the place of burial is later 
searched out by direction of M. de Sancy, the lost diamond is 
found in the dead man's stomach. 

Undimmed in this ghastly adventure, it rises from the grave 
to shine on the breast of Elizabeth of England (A.D. 1558-1603). 
From the last of the Tudors it passes to the Stuarts, and one of 
the few treasures that James the Second carries off in his flight 
from his throne (A.D. 1688) is the brilliant Sancy. Louis XIV. 
buys the gem from the king in exile (A.D. 1695), and it is held 
as one of the most precious of the crown jewels until the Revo- 
lution. In 1792 robbers break open the treasure chamber and 
bear it off" with other plunder. Again it is beyond tracing for 
years, till it reappears in the hands of a noble Russian family, 
the Demidoffs, from whom it passes to London merchants, and 
finally to the Maharajah of Puttiala. It may be that the adven- 
tures of two diamonds are fused in this tale, but it is none the 
less an outline of truth with the marvel of romance. 1 

Even Aladdin's wonderful palace, reared in a night by the 
hands of obedient genii, scarcely outstripped the glittering show 
of the court of the Great Moguls, enthroned in Delhi (A.D. 1526) 
by the arms of the Sultan Baber and his grandson Akbar, of the 

1 " A Treatise on Gems," Feuchtwanger. " Great Diamonds of the World," 
Streeter. 



26 THE DIAMOND MINES OF SOUTH AFRICA 

line of Timour the Tartar. Here embassies passed through the 
main gate of the palace along a magnificent avenue to the grand 
central square. Thousands of bodyguards in splendid dress lined 
the way, and behind the ranks richly caparisoned elephants 
were massed, waving flags of satin and silver. Dark eyes peered 
through the crimson hangings of the howdahs and the gilded 
lattices of the zenana cloisters bordering the square. Beyond 
the cloisters gardens outspread, with beds of lovely flowers and 
sheltering arbors and fountains splashing in sculptured basins. 

The entrance to the durbar or audience hall was through a 
pavilion hung with tapestries of purple and gold to a stately 
marble chamber, whose pillars and walls gleamed with rainbow 
hues. Under a canopy of flowered tissue on silver poles was set 
the imperial throne, the matchless triumph of Indian art. There 
strutted two peacocks fashioned deftly of jewels and gold to 
depict every plume and hue of the living creature. The out- 
spread tail seemed to flutter in mimicry of life with the sheen of 
sapphires and emeralds. The body was of enamelled gold and 
the eyes two radiant diamonds. Peacocks were emblems of the 
sun and of the descent of the Great Moguls from the sun through 
Chenghiz Khan. Ranged beside these splendid figures were 
stands bearing masses of unfading flowers, for every stem and 
leaf and petal was counterfeited in precious stones and metals. 

When the Great Mogul took his seat on his throne of solid 
gold studded with jewels, all bent low before his imperial 
majesty attired in cloth of gold blazing with precious stones 
in armlets and necklaces and crusted embroidery. Over the 
entrance to the hall was engraven in letters of gold: "If there 
be an elysium on earth, it is this." Here was at least a splendor 
of luxury beyond all rivalry. Never was shown, in vain Babylon, 
adventurous Tyre, or imperial Rome, any display as dazzling as 
the jewels of Delhi. 1 

'"The Turks in India," Henry George Keene, London, 1879. "His- 
tory of British India," Sir W. W. Hunter. Hunter's "Indian Empire." 
"Tales from Indian History," J. Talboys Wheeler. "Travels in the East," 
Vol. Ill, Forbes. 



THE ANCIENT ADAMAS 27 

Here the Koh-i-nur, Mountain of Light, sparkled, a price- 
less trophy. In the great battle of Pariput (April 2ist, 1526), 
when the last emperor of the Afghan-Lodi dynasty, Ibrahim, was 
beaten by Baber, the Rajah of Gwalior was " sent to hell," as 
Baber wrote grimly, and his most precious jewel valued "at 
half the daily expense of the whole world" came in tribute to 
Humaiun, the great sultan's favorite son. 1 Here, too, were the 
Koh-i-tur, Mountain of Sinai, and the Darya-i-nur, Sea of Light, 
and the Taj-e-mah, Crown of the Moon, and that prodigy of 
diamonds, the Great Mogul, presented to Shah Jehan by the 
Emir Jemla. 2 

These precious stones were coveted and hoarded with insane 
passion when every other lure in the boasted elysium was as 
Dead Sea fruit to the jaded senses. 
Shah Jehan, dethroned and impris- 
oned at Agra, sank to dotage, clasp- 
ing his casket of jewels, and trickling 
diamonds and rubies over his head 
and breast. When his son, Aurung- 
zeb, sent a messenger to borrow some 
of this hoard, the resentful old man 
threatened to break up the gems in 

. i i 11 i i i c \ i The Koh-i-nflr. (Present Cutting.) 

a mortar. Shah Kokh, the feeble son 

of Nadir Shah, who broke the peacock throne of the Moguls, 
was blinded by the Aga Mohammed in the vain effort to extort 
the Koh-i-nur. Then his head was shaved and circled with a 
ring of paste to hold boiling oil, but even this intensity of torture 
only forced the surrender of a ruby plucked from the crown of 
Aurungzeb. Shah Zaman, blinded by his brother Shuja, hid 
the Koh-i-nur defiantly for years in the plaster of his prison 
cell ; and Shuja, blinded by a third brother, Mahmud, yielded 
up the priceless stone to Runjeet Singh, only to save his family 
from agonizing death. 

In the sack of Delhi by Nadir Shah in 1739, the wonderful 

1 Memoirs of Sultan Baber. 

2 "Voyages en Turquie, en Perse et aux Indes," Tavernier, Paris, 1676. 




28 THE DIAMOND MINES OF SOUTH AFRICA 

store of jewels in the court of the Mogul emperors was borne away 
by the plunderers. It is supposed that the Great Mogul was 
broken at that time, and other famous diamonds were beyond 
tracing for years. The great gems were still more widely scat- 
tered upon the assassination of Nadir Shah, and some of the finest 
of the crown jewels of Europe have probably come from the 
hoards of Delhi. The Darya-i-nur and Taj-e-mah were set in a 
pair of bracelets which Sir John Malcolm saw at the court of 
Persia, 1 and they are still the most precious of the jewels of the 
Shah. Some have seen in the Orloff or Sceptre diamond of the 
Czar, the reappearance of the Great Mogul, but Streeter thinks 
that the Great Mogul has never come to light since the loot of 
the treasures of Nadir Shah by the Abdalli-Afghans. 

When the Koh-i-nur came into the hands of Runjeet Singh, 
he had the stone set in a bracelet which he wore proudly on 
every parade day. On his death-bed he sought to propitiate 
the gods by presenting this, the chief of his jewels, to the shrine 
of Jaga-nath (Juggernaut), but his hand was too weak to sign the 
warrant of delivery. So the gem descended to the young rajah 

Dhulip-Singh, and was held until the 
Indian mutiny and the seizure of the 
Punjaub by the English forces. Then 
the state property of the province was 
confiscated to pay debts due to the East 
India Company, but the Koh-i-nur was 
reserved for the English crown, and on 
June jd, 1850, this jewel, from earliest 
The Orioff. tradition the emblem of conquest, was 

placed in the hands of Queen Victoria by the messengers of 
Lord Dalhousie. 

Every precious stone of uncommon size has some adventure 
to tell, though its tale may not be a drama of as many acts as the 
Koh-i-nur' s career. What a strange story might be drawn from 
the Orloff of the sights in the temple of Mysore, when it was 
the eye of the Hindoo god, Sri-Ranga. 2 There was no other 
1 "Sketches of Persia," Sir John Malcolm, 1827. 2 Ibid. 




THE ANCIENT ADAMAS 29 

witness of the sacrilege of the French grenadier, masquerading as 
a devotee on the black and stormy night when he plucked out 
the precious stone eye and ran off through the British army lines 
to Madras. Here the captain of an English ship gave him 
.2,000 for his prize, but it cost Prince Orloff more than fifty 
times this sum when he bought it in Amsterdam to win back the 
favor of the Empress Catherine. 

The Regent lies in state, most lustrous and precious of the 
gems of the old French crown. The slave who found it buried 
in the bank of the Kistna River, A.D. 
1701, cut his leg deeply to pouch 
the stone in his flesh, and wrapped the 
wound in a thick bandage. At the 
first opening he ran away to the sea- 
coast and found refuge on an English 
merchant ship. But the lure of the 
big diamond was too tempting to 
the captain. When his ship was in the 
open sea, he flung the slave overboard 

to drown, and took the stolen diamond to sell to an Indian 
merchant, from whom it passed to the governor of Fort St. 
George, Thomas Pitt, grandfather of the great Earl of Chatham. 

It was one of the largest of all known diamonds, the rough 
stone weighing 410 carats, and Thomas Pitt would not suffer it 
to be out of his sight or touch day or night, though he was 
racked by the fear of thieves and murderers. While the alarm- 
ful gem was in his keeping, it is said that he never slept twice 
under the same roof, and moved from place to place in disguise, 
at a moment's caprice, to cover his tracks. Fortunately for his 
peace of mind, as well as his purse, he was able to sell his prize 
for ^135,000 to the Duke of Orleans, Regent of France in the 
minority of Louis XV. (A.D. 1715-1723). So the splendid stone 
made the fortune of the house of Pitt, and came to glitter in the 
most prodigal and luxurious court of Europe. It was held by 
the Bourbons until the French Revolution, and in 1792 it was 
stolen by the robbers who carried off the Sancy and thrown into 





30 THE DIAMOND MINES OF SOUTH AFRICA 

a ditch in the Champs Elysees. Here it was picked up with other 
plunder which the thieves did not dare to keep or offer for sale. 

Then it was uplifted again to the French 
crown and has held its place through 
revolutions that have unmade kings and 
emperors. 

So it might be told how " The Flor- 
entine " wandered from India through 
Tuscany to the Austrian crown, how 
the " Piggott " saw Clive's conquests 
(A.D. 1751-1767) and travelled to Eng- 
land with the governor of Madras and 

was crushed to powder by the dying Ali Pasha, how the 
" Star of the South " made its way from 
the sands of Brazil to glitter on the 
breast of the fantastic Gaikwak of Baroda 
while he killed disagreeable people with 
diamond dust, and how banished con- 
victs won their pardon from the Portu- 
guese crown by the discovery' of the 
Braganza, the largest diamond, if genu- 
ine, that the world ever saw. 1 ThePiggott. 

No one can say of a true diamond 
story, " it is closed " ; for diamonds 
outlast dynasties, and their wander- 
ings may be on the verge of renewal 
when they seem to be ended. " A 
jewel may rest on an English lady's 
arm that saw Alaric sack Rome, and 
beheld before what not ? The 

The Star of the South. 

'"Great Diamonds of the World," Streeter. " Diamonds," W. Pole, 
London Archzological Trans. , London, 1861. "Diamonds and Precious Stones," 
H. Emanuel, London, 1865. "Outlines of Mineralogy," J. Kidd, Oxford. 
"Traite Complet des Pierres precieuses," Charles Barbot, Paris, 1838. "The 
People of India," J. Forbes Watson and J. W. Kaye, Editors. " Gemmarum et 
Lapidum Historia," etc., Boetius, 1647. 





THE ANCIENT ADAM AS 31 

treasures of the palaces of the Pharaohs and of Darius, or the 
camp of the Ptolemies, come into Europe on the neck of a 
vulgar pro-consul's wife to glitter at every gladiator's butchery 
at the amphitheatre ; then pass in a Gothic ox-wagon to an Arab 
seraglio at Seville ; and so back to its native India, to figure in 
the peacock throne of the Great Mogul; to be bought by an 
Armenian for a few rupees from an English soldier, and so, at 
last, come hither." 

The illustrations of the historic diamonds shown in this chapter have been 
made from photographs of facsimiles of the stones, and are the exact sizes of the 
originals. 



CHAPTER II 



IN TRADITIONAL OPHIR LAND 




CHILD picking a shining pebble for a play- 
thing from the gravel edging a river was this 
sport of blind chance the revelation of the mar- 
vellous diamond fields of Africa? In narrow 
fact, yes ; but in a wider, truer range of view, 
this discovery was the crown that sooner or 
later must reward the search of daring adventurers and the push 
of stubborn pioneers into the dark heart of the continent. 

There was no chance in the strain of pluck that braved 
strange perils to reach traditional Ophir and the pits of King 
Solomon's mines, that wandered far in quest of the golden cities 
of Monomotapa, that tore the wilderness from the clutch of the 
lion and vulture, and beat back the frantic impis of Tchaka, 
Dingaan, and Umsilikazi. The ardor and the toil and the 
courage and the blood of ten generations of explorers were 
spent before it was possible for a little child to play pitch and 
toss with the pebbles of the Orange River and clasp a rough 
diamond in his heedless hand. 

Two dominant motives were fused with the high-spirited zeal 
for exploration that so signally stamped the fifteenth century, 
the opening of an all-sea route to the Indies, and the grasp of 
the riches of lands behind the veil. In the unknown there is 
space for any vault of fancy, and in that romantic age her soaring 
wings were rarely clipped. One may be moved to smile at the 
fantastic visions of the men who found the southern waterway 
to the Indies, and added a new world to the old ; but there will 
be no sneer in the smile of any one who can measure his own 
debt to experience, and put himself back five centuries to stand 

3 2 




: _-SlT < 

or 



IN TRADITIONAL OPHIR LAND 33 

on the deck with Cam, Bias, and Da Gama, or the still more 
greatly daring Columbus. 




Visscher's Map of Africa, 1662. (From the original in the British Museum.) 

But who can to-day feel the hopes and fears that shook 
those strong hearts ? Who can lay the course for their clumsy 
caravels over the unknown stretches of ocean ? Who can sail 
on with them day after day and night after night without a chart 



34 



THE DIAMOND MINES OF SOUTH AFRICA 



or buoy or beacon or surf-rocked bell ? Who can start from 
fitful sleep to pierce the night with straining eyes or watch for 
the glimmer of the dawn on sea-girt horizons ? Who can recall 
their racking fears or the dazzling images ever forming and dis- 




Africa, from an early Dutch Map. 



solving in the alembic of their fancy ? With every daybreak 
the isles of Atlantis might spring into view, or gardens fairer 
than the golden Hesperides, or monsters more horrific than 
dragons, guarding hoards beyond the dreams of avarice, or, per- 



IN TRADITIONAL OPHIR LAND 35 

chance, even the realms of some potentate accustomed to make 
footstools of princes with stiffer necks than haughty Xerxes or 
the terrible Tamburlane. 

Amid the drift of such cloudy conceits there was one more 
clearly shaped and persistent than the rest. Somewhere below 
the equator, in the unknown expanse of Africa, tradition placed 
the home of the Queen of Sheba, King Solomon's mines, and the 
marvels of Ophir. Every adventurer skirting the South African 
coast hoped to touch with certainty the shore of this delectable 
country. The alluring recital in Kings and Chronicles glittered 
before his eyes. 1 In fancy he saw the gathering of the ships in 
" Ezion-Geber, which is beside Eloth on the shore of the Red 
Sea," and how this fleet came to Ophir and fetched from thence 
gold, four hundred and twenty talents, and brought it to King 
Solomon. He saw, too, the coming of the Queen of Sheba to 
the king to prove him with hard questions, and the great train 
that followed her with camels that bare spices and very much 
gold and precious stones. Then it was told him how the queen 
was overcome by Solomon's wisdom and grandeur until " there 
was no more spirit in her," and she gave the king one hundred 
and twenty talents of gold, and of spices very great store, and 
precious stones. Following this tribute came the regular flow, 
from Ophir to Judea, of gold and gems and almug trees in the 
transports of Tyre. With such a fountain of supply, it was 
easy to credit the wonderful tale of the targets and shields of 
beaten gold, of the throne of ivory overlaid with gold, and of all 
the other displays of Solomon's splendor. If the king's gold 
made silver to be in Jerusalem as stones in the eyes of the 
chronicler, it is not surprising that this vision came down 
undimmed to the days of Da Gama. 

But how to find the source of this flow was the puzzle that 
faced the explorer. Unfortunately the old chroniclers had 
omitted to give any landmarks of King Solomon's mines. Sur- 
mise strayed down the eastern coast of Africa, and the close 
commercial connection between southwestern Arabia and the 
1 1 Kings ix., x ; z Chronicles viii., ix. 



36 THE DIAMOND MINES OF SOUTH AFRICA 




Outline Copy of the Catalan 
(In the original the shore line has almost illegible names, 



equatorial coast region of East Africa was unquestionable. 
Herodotus declares that East Africa at its furthest known limits 
supplied gold in great plenty as well as huge elephants and 
ebony. The Alexandrian geographers mark rudely the East 
African coast line to Zanzibar, and attest the relations between 



IN TRADITIONAL OPHIR LAND 



37 




Mappermonde, 1375. 

which, for the sake of clearness, have been omitted here.) 

this coast and Arabia Felix. Eratosthenes observes that naviga- 
tion extends down East Africa beyond Bab-el-Mandeb, " along 
the myrrh country, south and east as far as the Cinnamon coun- 
try, about five thousand stadia." Ptolemy, in the second cen- 
1 Strabo, XVI, Chap. IV, 4. 



38 THE DIAMOND MINES OF SOUTH AFRICA 

tury A.D., describes quite accurately the east coast of Africa as 
far as Zanzibar and Ras Mamba Mku. His information was 
chiefly derived from Arabian merchants. But, as Schlechter has 
closely pointed out in his admirable monograph, 1 there is no 
trace or hint anywhere during the Greek and Roman periods of 
antiquity of any colony or emporium south of the Zanzibar 




Outline Copy of the Portolano Laurenziano, 1351. 

coast, and not long after the time of Herodotus the gold im- 
ports of Arabia had shrunk, to inconsiderable importance. With 
the decline of the Himyaritic Kingdom in Arabia, soon after the 
second century of our era, there was a falling off of commercial 
enterprise and intercourse with Africa, so marked that even the 

1 " Periplus of the Erythraean Sea," Henry Schlechter, The Geographical 
Journal of the Royal Geographical Society, July, 1893. 



IN TRADITIONAL OPHIR LAND 



39 



notable map of the Arabian Edrisi, in 1154 A.D., shows how 
slight and vague was the advance in the knowledge of the Dark 
Continent from the days of the Alexandrian geographers. Still 
this old chart gives some substantial proof of the communica- 
tion of Arabian traders with the natives on the East African 
coast. But on this map the African coast appears to curve 




Africa de Mappermonde, Juan de la Cosa, 1500. (This map was made only fourteen years after 
the discovery of the Cape of Good Hope, and is one of the earliest known maps giving the 
entire contour of Africa with approximate accuracy.) 

east continuously from the mouth of the Red Sea, and Edrisi 
was plainly ignorant of the abrupt trend to the south from Cape 
Guar-da-fui. Yet he shows rudely the islands lying off the east 
coast of Africa, and, south of Sokotra, traces the African main- 
land in three divisions, Zendj (Zanzibar), Sofala, and Vakvak. 

With all its imperfections this Arabian map was in advance 
of any European portrayal of South Africa. It was the prevail- 
ing belief in the Middle Ages, " bequeathed from antiquity," as 



40 THE DIAMOND MINES OF SOUTH AFRICA 



Justin Winsor observes, that " owing to the impassable heats of 
the torrid zone, it could not be discovered whether this region 
were inhabited or whether land existed there." Map makers 
plainly made the bounds of land and water beyond the equator 
from sheer surmise, and the confession was commonly frank that 
the land was terra incognita and the ocean a sea of darkness. 
" Most famous of all these early maps" (of the Atlantic Ocean), 




Dutch Ship of the XVIIIth Century. 

says Winsor, 1 " was the Catalan Mappermonde of 1375." It was 
probably the one best known by the sailors sent out by Prince 
Henry of Portugal, in the year 1413, to follow down the Atlan- 
tic shore line of Africa. On this map, all known Africa is 
bounded on the south by a line drawn eastward from Finisterra, 
off the mouth of the Rio Del Oro, about 23 north of the 
equator, nearly across the continent to the Egyptian Nile. In 
the Portolano Laurenziano of 1351, the outline of Africa is given 

1 " Narrative and Critical History of America," Vol. I, p. 55, Justin Winsor. 



IN TRADITIONAL OPHIR LAND 




Dutch Ships of the XVI Ith Century. 

an approach to reality that is highly remarkable, but it is clearly 
a happy stretch of guesswork. 1 

All of the region south of Cape Non was practically un- 
known to the adventurers of the fifteenth century. 2 Their ears 
were filled with doleful tales of the calms and storms, the 




Dutch Ships of the XVIIth Century. 

1 " Life of Prince Henry of Portugal, surnamed the Navigator, and its Results," 
R. H. Major, London, 1868. 

2 Chief of the charts in the fifteenth century were those of Andrea Bianco, 
"Atlas," 1436, and " Carta Nautica." Justin Winsor, "Narrative and Critical 
History of America," Vol. I, p. 55. 



42 THE DIAMOND MINES OF SOUTH AFRICA 

mud-banks and the fogs, of the Sea of Darkness. If by any 
stretch of daring they might cross the equatorial line, they were 




Dutch Ship of the XVI Ith Century. 



burdened with the fear that they would begin to slide down an 
inclined plane with a rush that would pitch them headlong into 




Dutch Ships of the XVI Ith Century. 



some bottomless abyss. The only assurance of a happier issue 
was the bare tale of old Herodotus of some nameless Phoenician 



IN TRADITIONAL OPHIR LAND 



43 



sailors who had skirted the coast south from the Red Sea in the 
days of Pharaoh Necho (610-594 B.C.), and returned nearly 
three years later through the Pillars of Hercules and the Medi- 
terranean. These sailors brought back, with their load of ivory, 
feathers, and gold, the report that during a considerable part of 
this voyage they had the sun on their right hand. It is this 
detail that now chiefly confirms the story, but this was beyond 
the credence of Herodotus, 1 and it would seem that this ancient 
mariner's tale was soon generally disbelieved, for the special 




Dutch Ships of the XVI Ith Century. 

searches made in the Alexandrian library by Eratosthenes and 
Marinus of Tyre in the third and second centuries B.C. brought 
to light no other records or traces of the voyage. So it was not 
with reliance on this alleged circumnavigation that the adven- 
turers of Portugal groped painfully for seventy years along the 
coast, until the daring Dias set his stone crosses at Angra 
Pequena and Algoa Bay and sighted the turning point of the 
path to the Indies in the frowning Cabo de Todos los Tor- 
mentos. King John was quick to see the promise in the land 
1 " Herodotus," Bk. 4, 42, Rawlinson. 



44 THE DIAMOND MINES OF SOUTH AFRICA 

of Dias and change the Cape of Storms to Cabo de Boa Espe- 
ranza, but ten years passed before Vasco da Gama followed 
down the trail and rounded the Cape in the immortal voyage 
that reached the long-sought Indies six years after Columbus 
had touched the island hem of the new world. 1 




Dutch Ships of the XVIIIth Century. 

The completed circling of Africa by European adventurers 
was a no less memorable achievement of Da Gama. He touched 
at Mozambique on the first of March, 1498, and there saw gold, 
in the hands of Arabs, that had passed up the coast from Sofala. 
Nearly twenty years before, a Portuguese courtier, Pedro de 
Covilhao, had reached Sofala in an attempt to pass to India by 
way of Egypt. 2 

For many years and possibly for many centuries there had 
been a trickle of gold from Sofala through Arab traders, and 
Da Gama saw enough of it to move his king to lay his hands 
upon it. In the expedition of Cabral, which followed in the 
wake of Da Gama in 1500, the great captain, Bartholemeu 

1 " Prince Henry the Navigator," C. Raymond Beazley. 

2 "The Portuguese in South Africa," George McCall Theal. " South Africa 
from Arab Domination to British Rule," R. W. Murray, editor, London, 1891. 



IN TRADITIONAL OPHIR LAND 45 

Bias, was specially commissioned to seek the source of the 
gold stream. Bias was drowned in the storm which sunk four 
ships of this fleet, but Cabral took a vessel carrying gold from 
Sofala and sailed to Kilwa, where the Arab Ibrahim and his 
forefathers had been drawing gold from Sofala for a long term 
of years. Upon the report of Cabral, Ba Gama turned out of 
his way to Mozambique in his second voyage, in 1502, to enter 
Sofala and take possession of Kilwa, and three years later Pedro 
da Nhaya sailed from Lisbon with six ships and built a fort 
and trading station at Sofala. 

Behind this persistent push to Sofala there was more than 
the actual showing of gold. Here was one of the traditional 
gateways to King Solomon's mines, and the Portuguese were 
quick to embrace the tradition. They gave the glittering name 
of Ophir to their fort. South of the fort there runs a river, 
called by the Arabs Sabi, and this was pounced upon as a 




Dutch Ships of the XVIIIth Century. 

probable twist of the old Hebrew Sheba. From those days 
Fort Ophir was the starting point of Portuguese adventurers 
in search of the fountain head of Solomon's treasures. 

The Portuguese then had uncommonly sturdy sea-legs and 
asked nobody to show them the way over the ocean foam, but 



46 THE DIAMOND MINES OF SOUTH AFRICA 




(f. 



o 
u 



o 

X 



H 



Q 

O 
E 

s 



o 

C/3 

o; 
u 




ry 

i v ^- 



IN TRADITIONAL OPHIR LAND 



47 



they were far less ready to weary their legs with trudging over 
mountain ridges or scrambling through the dense thickets of 
the rugged land west of Sofala. The Arab traders were more 
ready to venture inland, but there is no evidence to show that 
any of them went farther than a few hundred miles, at most, 
from the seacoast. It was an exceedingly difficult country to 
penetrate, and the savage natives were jealous of any approach, 
if they did not stubbornly bar the way and murder intruders. 



E B JI A N 
EAST 
F E I C . 




Macloul 
mm 
BECHUANALAND 



Map showing the Position of Ancient Ruins in Rhodesia. 



The horrid death of the first Portuguese viceroy was a warn- 
ing that struck deep into the hearts of the earlier adventurers. 
Francisco d' Almeida, returning with his fleet from India in 
1510, touched the African coast near the first landing of Diaz. 
To resent some little clash with the nearest native tribe he led 
a troop of soldiers inland to surprise their village, but was way- 
laid in the bush and his troop was put to flight by a hail of 
darts and stones. D' Almeida put his ensign in the hand of a 
trusty follower, but in the next moment he was stabbed in the 
throat by an assagai and his head was crushed by the swing 



4 8 



THE DIAMOND MINES OF SOUTH AFRICA 



of a knob kerrie. Sixty-five of his picked swordsmen fell with 
him and the rest only saved their lives by abject flight, chased 
to the shore by a little band of naked negro dwarfs. 

This was the greeting of a weak and puny coast tribe. What 
then might be feared from the rallying of the fierce and stalwart 
blacks of the Bantu tribes, under some ruthless chief, in the 
fastnesses of the mountain land encircling the gold of Ophir ? 




Insiza Ruins. 



Still there was an enticing trickle of gold dust and nuggets from 
inland mines to Sofala, and the flow of resplendent stories was 
vastly bigger than the golden stream in sight. So in 1569 it 
was resolved to make an extraordinary effort to penetrate to the 
source of the gold. The East African coast was placed under 
command of a governor independent of the viceroy of India. 
Francisco Barreto was made the first governor, with instructions 
to raise a force of a thousand men and lead them on to the 
capture of Ophir. The young cavaliers of Lisbon flocked 
eagerly to Barreto's standard. He led the way up the Zambesi 
with a high-spirited troop, but the gay soldiers were soon 



IN TRADITIONAL OPHIR LAND 



49 



scorched by the sun, torn by thorns, and cast down by fevers. 
The Kalangu tribe was then the strongest of any living between 
the Sabi and Zambesi, and Barreto sought to win the good will of 
its head chief by offering to beat his rival. This offer made him 




Insiza Ruins. 



welcome, and he kept his promise, but he was soon after obliged 
to appoint Vasco Fernandez Homem to the command of his 
troop and to return to the coast. Homem soon followed him 
with the dispirited remnants of the adventurers. Barreto did 
not live to see the return of his broken expedition, and Homem 



50 THE DIAMOND MINES OF SOUTH AFRICA 

succeeded him as governor. Then the new governor tried an- 
other way of approach to the gold field, and finally pushed a party 
through from Sofala to the foot of the mountain which the 
Kalangu tribe called Kara and the Arabs Aufur, transmuted 
forms, it was thought, of the Hebrew Ophir. Near the base 
of this mountain were placers yielding nuggets worth from two 




Insiza Ruins. 



to three thousand dollars, but the ordinary toil of placer wash- 
ing was so disgusting to the Portuguese visionaries that they 
gloomily turned their backs on the mines of Abasia and the 
rock mark of Ophir and wearily made their way back to Sofala. 1 
This disappointment dulled the glitter of some old stories, but 
there were plenty of new ones to dazzle men's minds. 

It is likely that the most accurate, as it certainly is the full- 
est extant, account of the mining in Ophir land is given in the 
story of the old Spanish author, Joano de Barros, whose life 
spans the first three quarters of the sixteenth century. 2 It is 
too much to expect that his " Da Asia " should be free from 
the coloring of the ardent fancy and the myths of the age, but 
underlying his narrative there is, at most points, a credible 
basis of personal observation and the current reports of many 
witnesses. He held several high offices in the Indian and 

1 "The Portuguese in South Africa," Theal. " Conferencias Celebradas na 
Academia Real das Sciencias de Lisboa, Acerca dos Descobrimentos e Colonisa- 
c,oes dos Portuguezes na Africa." [At Lisbon, 1892.] 

2 "Da Asia," Joano de Barros (1496-1570). 



IN TRADITIONAL OPHIR LAND 51 

African establishments of Portugal, and had exceptional oppor- 
tunities for preparing his remarkable memorial. 

In his description the "mines of Manica" are placed "some 
fifty leagues west of Sofala." The Portuguese league was 3.84 
English miles, and De Barros was as loose as contemporary writ- 
ers in the measure of distances. " All gold found there is in 
dust," he writes, " and the workers have to carry the earth 
which they dig to some place where water can be had. Nobody 
digs more than six to seven spans deep (four to six feet), and if 
they go to twenty, they come to hard rock." 

Beyond the Manica placers, in positions not defined, were 
the mines of Boro and Quiticui. There nuggets were found 




Khami Ruins. 



" embedded in reefs some already cleared by the winter tor- 
rents ; hence, in some of the pools, such as remain in summer, 
the miners dig down and find much gold in the mud brought 
up. In other localities, where are some lagoons, two hundred 
men set at work to drain off about half the water, and in the 
mud which they sift they also find gold, and so rich is the 



52 THE DIAMOND MINES OF SOUTH AFRICA 




ground, that if the people were industrious, great quantities 
could be had ; but they are so indolent that stress of hunger 
alone will keep them at work. Hence Moors (Arabs) who 

visit those districts 
have recourse to a 
ruse to make them 
diligent. They cover 
the negro men and 
women with clothes, 
beads, and trinkets in 
which they delight, 
and when all are 
pleased trust every- 
thing to them, telling 

Khami Ruins. fa em tQ go tQ worlc 

the mines, and on their return, they can pay for those advances ; 
so that in this way, by giving them credit, they oblige them to 
work, and so truthful are the negroes that they keep their word- 
" Other mines lie in the district called Toroa, ruled by a vas- 
sal of Benomotapa. These are the oldest known in that region. 
They are in a plain, in 
the middle of which 
stands a square fortress, 
all of dressed stones, 
within and without, well 
wrought and of marvel- 
lous size, without any 
lime showing the join- 
ings. The walls of this 
fortress are over twenty- 
five spans high (18 to 19 
feet) but the height is not so great compared with the thickness. 
And above the gateway of that stronghold there is an inscrip- 
tion which some learned Moorish traders who were there could 
not read nor say what writing it was. And around this build- 
ing are others on some heights, like it in the stonework, in 







Gold Ornaments found in Ancient Ruins. 



IN TRADITIONAL OPHIR LAND 



53 



which is a tower twelve bracas (72 feet) high. All these struc- 
tures the people of the country call Symbaoe, which with them 
means a royal residence. They stand west of Sofala, under 
latitude 20 and 21 south, one hundred and seventy leagues 
more or less in a straight line. ... In the opinion of the 
Moors who saw them, they seemed to be very ancient and were 




Khami Ruins. 



built there to hold possession of those mines, which are very old, 
from which for years no gold has been taken owing to the wars." 
The latitude and position of the Symbaoe of De Barros cor- 
respond closely with the site of the ruins of Zimbabwe, described 
three hundred years later by the explorer Karl Mauch. Both 
Zimbabwe and its antique form, Symbaoe, are plainly versions 
of the local Bantu nzimba-mbuie, a house of the chief. It is 
true that the Zimbabwe of Mauch is only two hundred and forty 
miles west of Sofala, but the leagues of the old chroniclers were 
not laid off with the tape line. 



54 THE DIAMOND MINES OF SOUTH AFRICA 

Who was this Benomotapa whose vassal was housed in such a 
castle ? the mighty black sovereign of whom Camoens sings 

" Ve do Benomotapa o grande imperio, 
De Salvatica gente, negra e nua " ? 

In dull fact Benomotapa was simply the corrupted plural 
form of Monomotapa, signifying Lord of the Mountain, or by 
a possible stretch of derivation, Master of the Mines. 1 This 
was one of the hereditary titles of the head chief of the Kalangu 




Khami Ruins. 

tribe, the largest and strongest of any then living between the 
Sabi and Zambesi. His dwelling was at the foot of Mount 

1 "The Portuguese in South Africa," George McCall Theal. 

Bent says the name Monomotapa should be written Muene-matapa, or " lord 
of Matapa," simply "a dynastic name, just as every petty chief in Mashonaland 
to-day has his dynastic name, which he takes on succeeding to the chiefdom." 
"The Ruined Cities of Mashonaland," p. 285. Both titles have in fact the same 
meaning : the first components bena and mono being the still current Bantu words 
bwana, bana, muene, mwana, that is 'lord,' 'master,' 'chief,' 'ruler.' The 
second part, motapa, common to both, probably means a mine, from the Bantu 
word ta/>a='to dig,' 'excavate.' "Africa," Vol. II, p. 372. (Stanford's 
Compendium.) A. H. Keane. 



IN TRADITIONAL OPHIR LAND 



55 



Aufur, which was held in such traditional reverence that the 
chief would not permit the Portuguese to ascend it. There was 
nothing of imposing 
splendor in the huts 
of the chief who re- 
ceived the embassy 
of Francisco Barreto, 
but no lack of evi- 
dence could prevent 
romance from creat- 
ing an African em- 
pire under the sway 
of Monomotapa. 
Some corner-stones 
for this structure were 
found in the remains 
of the works of a 
people of far higher 
civilization than any 
of the existing native 
tribes, and these relics 
were prizes to a fancy 
that clutched greedily at every drifting straw of report, tradition, 
and myth supplied by Arabs and negroes. 

Every one in the suc- 
cession of romancers, in 
the sober cloak of histo- 
rians, of South Africa 
would outdo his forerun- 
ners in inflating the bal- 
loon of the traditional 
empire. The old Dutch 
writer, Kleveer, finally 
puffed it up to the bursting limit by bounding it " on the 
east, south, and west by the Atlantic, and north by the king- 
doms of Congo, Abyssinia, and the Zanzibar country. Even 




Zimbabwe Ruins. 




Zimbabwe Ruins. 



56 THE DIAMOND MINES OF SOUTH AFRICA 

Dapper, 1 whose really great work is by far the most important, 
comprehensive, and creditable presentation of the Africa of the 
seventeenth century, jots down gravely most fantastic details of 
the empire ruled by the royal line of Monomotapa. He paints 
a mammoth palace with four grand gateways leading to a succes- 
sion of halls and chambers, rivalling the handiwork of the slaves 
of the lamp of Aladdin. All the ceilings of the rooms were gilt 
or covered with golden plates. For the furnishing of sumptuous 




A 



Zimbabwe Ruins. 



couches and chairs there was gilding and painting in rainbow hues 
and artful inlaying with enamel. Ivory chandeliers, hanging on 
silver chains, filled these resplendent halls with light. When his 
majesty deigned to rise from his imperial bed, he was clothed by 
his valets in garments of native silk. All his servants approached 
him on bended knees and served him like dumb slaves. His 
table service of the finest porcelain was decorated with wreaths of 
gold, cunningly wrought in the fantastic forms of natural coral. 

1 " Naukeurige Beschrijringe der Afrikaensche Gewesten," etc., Dr. O. Dap- 
per, Amsterdam, 1668. 



IN TRADITIONAL OPHIR LAND 




Zimbabwe Ruins. 



Two pounds of gold was daily spent in perfume for the royal 
nose, and torches of incense flamed day and night around him. 
When he took an airing, he was borne in a gorgeous palanquin 
on the shoulders of four of his trembling nobles, and his head 
was shielded from the profaning sun by a canopy studded with 
precious stones. If he was impatient of this slow promenade, he 




Zimbabwe Ruins. 



THE DIAMOND MINES OF SOUTH AFRICA 



might mount on an elephant's back, but on nothing meaner, for 
nobody in that wonderful country would ride on any other animal. 
It is small wonder that the court of monarchs of this splen- 
dor, and their golden cities of Davaque and Vigiti Magna, were 
ardently hunted for by adventurers, thirsty for every romance 
gilding the dismal stretches of sand and thickets and rocks 
which encircled them with the threads of a trail to the glittering 
realm of Monomotapa. But the expeditions of Barreto and 

Homem were so painful, costly, 
and discouraging that for many 
years no more explorations were 
undertaken by the Portuguese 
crown. The spirit of chivalric 
adventure drooped low after the 
gallant young king Sebastian 
fell in battle with the Moors 
in 1578, and even the spirit 
that had so greatly spread the 
commerce of Portugal was los- 
ing its vigor. There was a 

- r "%1&?!^Hl^^HHH3l momentary arousal in the be- 
ginning of the seventeenth cen- 
tury, when some rich silver ore 
was sent to Lisbon by the 
governor of Mozambique. It 

^^^ was believed that this ore came 

f //'i . tEgfcv^^. ^feflBBI^BBlBB from veins in a region called the 
Zimbabwe Ruins. Kingdom of Chicova, stretch- 

ing north from the bank of the Zambesi ; but there was no 
definite report of the location. Still there was such an impulse 
in the sight of this silver that the order was sent to despatch 
five hundred soldiers to Chicova. No such force could be 
mustered, but Nuno Alvares Pereira set out from Mozambique 
with a hundred men. Soon Pereira was the victim of jeal- 
ous maligning, and was superseded in his command by Diogo 
Sinoes Madeira. This commander succeeded in placing a few 




IN TRADITIONAL OPHIR LAND 



59 



trading stations along the Zambesi, and made a pretence of 
opening mines by shipping some little silver to Portugal ; but 




Zimbabwe Ruins. 

after a dozen years of costly maintenance, it was shown by the 
search of Pereira that the pretended discovery of silver was a 




Zimbabwe Ruins. 

fraud, and disgusted Portugal abandoned the enterprise in I622. 1 
1 "The Portuguese in South Africa," Theal. 



60 THE DIAMOND MINES OF SOUTH AFRICA 



From that year nothing of note was attempted from the 
stretch of seaboard loosely held by a few feeble garrisons. 
Beyond the vague traditions and romances there were no guide- 
books to the rich realm of any African monarch, and there was 
no point on the South African coast outside of the Portuguese 
strip where the least enticement was shown to any visiting ship. 
Nowhere was there any evidence of an approach to civiliza- 
tion, and there was not even the gilding of barbarism. The 

shore tribes 
were filthy, 
famine-hunted 
negroes, who 
had, at most, a 
little ivory or 
a handful of 
feathers to bar- 
ter for trinkets. 
There was an 
intermixture of 
blood and a 
medley of 
tribes and tribal 
names that 

confounds any tracing of distinction beyond a few blurred divi- 
sional lines. 

When the Dutch and English began to tread upon the heels 
of the Portuguese in Africa, in the opening years of the seven- 
teenth century, the tribes of the extreme south and along the 
southwesterly Atlantic coast might be roughly grouped under the 
name of Hottentots, or, as they called themselves with monstrous 
conceit, Kwa-Kwa, men of men. In this assertion there is 
plainly to be seen the origin of the Arabic Vakvak, the name 
sketched in by Edrisi on his map beyond Sofala. The south- 
east African coast was held by tribes of the wide-spreading 
Bantu family, lumped together by the Arabs as Kafirs. Filtered 
in between the Bantus and Hottentots were the pigmy Sana, 




Zimbabwe Ruins. 



IN TRADITIONAL OPHIR LAND 



61 



rudely bunched as Bushmen. 1 There was endless wrangling 
and fighting among the tribes, regardless of any common flow 
of blood, and the Bantus and Hottentots were continually clash- 
ing like wildcats. Their only union was in their hate of the 
Bushmen, who were hunted 
from cover to cover, to hide 
in crevices in the rocks or in 
holes in the desert sand, from 
which they might sally, wasp- 
like, with the deadly sting 
of their poison-tipped arrows. 

In view of the repulsive 
face of the South African 
coast lands it is not surpris- 
ing that Francis Drake and 
many other bold voyagers 
circled the Cape of Good 
Hope without landing to 
seek for traditional treasures. 
But with the opening of the 
seventeenth century, Table 
Bay became a regular stop- 
ping place and refitting station for the ships of the English East 
India Company. For twenty years this slight hold on the con- 
tinent was maintained, but it was so lightly prized that it was 
dropped in 1620 by a shift of the station to St. Helena. Thirty- 
two years later the Dutch East India Company took formal 
possession of the Cape and its adjoining bay without any chal- 
lenging protest, and built their fort Good Hope as the first 
stronghold of the Dutch dominion in southern Africa. With 
this foundation the search for the golden realm of Monomotapa 
was vigorously and persistently revived. 

Jan van Riebeeck, the leader of the Dutch colonizing expe- 
dition and the first commandant of the fort and settlement at 




The Old East India House, Leadenhall Street, 
London. 



1 "South Africa," George McCall Theal, London, 1888-1893. 
African Tribes," Sutherland. 



South 



62 THE DIAMOND MINES OF SOUTH AFRICA 

Cape Town, was a man of ardent spirit and uncommon energy. 
He had entered the company's service as a surgeon's assistant, 
but his ambition and ability had soon pushed him to the front 
and marked him as a man to fix and strengthen the grip of the 
great trading company on the turning-point of the way to the 
Indies. In his portrait dark, sanguine eyes are set under a 
high, full forehead, crowned with thick waving hair of a cavalier 
cut, in keeping with his trim mustache. His well-moulded 
features and resolute chin have the stamp of refinement as well 
as action. He quickly put his hand to every practical device to 
make the new settlement productive and self-supporting. Nine 
months after his landing the first crop of wheat was reaped at 
the Cape. In the following year he set out vines from the 
Rhine. In his own vineyard the muscatel grape grew luxuri- 




The Landing of van Riebeeck. 

antly, and a few years later he made the first Cape wine, a high- 
flavored Constantia. In the same year, 1658, maize was brought 
to the colony from the coast of Guinea and successfully planted. 
To the introduction of the olive, particularly urged by the direc- 
tors of his company, he gave unremitting pains, and succeeded 
in rearing a fine grove of fruitful trees on his own plantation at 
Wynberg. In his stretch of experiment he even tamed young 
ostriches and stocked the neighboring islands with rabbits. 1 

Such a man was not likely to be heedless of the chances for 

'"South Africa," Theal. "On Veld and Farm," Frances MacNab, 
London, 1897. 



IN TRADITIONAL OPHIR LAND 



the possible enrichment of his company by penetrating to the 
seat of the traditional empire and possibly to King Solomon's 
mines. He reckoned that, in any event, his exploring parties 
would be likely to succeed in uncovering ore beds of some use- 
ful metal, if not of gold and silver. But he seems to have had 
great confidence in the traditions of Monomotapa, and it is 
known that he had before him the highly colored work of 
the Dutch traveller and author, Linschoten, as well as current 





Portrait of Johan Antonyse van Riebeeck. 
First Commandant of the Cape of Good 
Hope. Born 1618, died January 18, 
1672. 



Portrait of Maria de la Querellerie of Que- 
rellerius, Wife of Johan Antonyse van 
Riebeeck. Born October 28, 1629, died 
November 2, 1664. 



Portuguese books infused with the romance of Africa. His 
calculation plotted the location of Davaque, the chief seat of 
the splendors of Monomotapa, at a point 828 miles N.E. of 
the Cape of Good Hope, and 322 miles W. from the Indian 
Ocean, curiously near the present Witwatersrand. Davaque 
was built by tradition on the banks of the river Spirito Sanctu, 
flowing into the Indian Ocean at Delagoa Bay. Nearer still 
to the Cape was another El Dorado, the city of Vigiti Magna, 
which was confidently located on or near the meridian of 30 S., 
and not much more than three hundred miles from the Cape. 



64 THE DIAMOND MINES OF SOUTH AFRICA 

The first push into the unknown land north of Fort Good 
Hope was made in 1657 by a little party headed by Abraham 
Gabbema, Fiscal, and Secretary of the Council of the colony. 
Gabbema led the way to the first big beacon in sight, a peak 
with a grotesque flat top which the colonists had already chris- 
tened Klapnuits, or night cap mountain. Skirting the base of 
this peak he pushed to the next conspicuous landmark, bearing 
toward the west, a mountain with bare rugged pinnacles of rock, 
which the explorers dully called Great Berg, and gave the same 
name to the river flowing below. 




Constantly. 



It was in the middle of October when the party set out, but 
this was the prime of the springtime in South Africa. On the 
lower slopes of the Great Berg herds were grazing that had 
never seen the face of a white man nor felt the sting of a bullet. 
Zebras capered over the hillsides, the unwieldy rhinoceros wal- 
lowed in the high grass, and hippopotami plunged and snorted 
in the turbid rivers. Every step of the way was a new wonder- 
ment to the explorers, and when the rising sun struck the moun- 
tain tops with its flame, two transfigured peaks gleamed like 
prodigious gems in their eyes, and were forthwith distinguished 



IN TRADITIONAL OPHIR LAND 65 

as Paarl and Diamant. These sunlit crests were the only things 
in sight, however, that had any glitter of the realm of Mono- 
motapa, and after a little further advance into the unknown field, 
Gabbema's party turned back. 

The next excursion was more daring. By promising rich 
rewards van Riebeeck formed a party of thirty volunteers headed 
by Jan Danckert. They took along a small stock of bread on 
three pack oxen, relying for their main supply of food on the game 
which they might kill on their way. These hardy volunteers 
plodded north, inclining to the west along the foot of the coast 
range. They saw whirlwinds of dust and a few roving Bushmen, 
but nowhere any trace of a monarchy except what they called " A 
Kingdom of Moles," where the burrowed ground sank under 
their feet and they could hardly flounder along. In December 
they reached a river flowing toward the Atlantic, on whose far- 
ther shore they saw a herd of more than two hundred elephants 
feeding. So they called the stream Olifants River, a name 
which it has borne since that day, and trudged back wearily to 
tell their story to the commandant at the Cape. Within ten 
days after their return, January 20, 1661, van Riebeeck, the un- 
tiring, mustered another party, of thirteen adventurers and two 
Hottentot attendants, and sent them away on the track of the 
discoverers of Olifants River. 

Corporal Pieter Cruythof led ofF this party, which succeeded 
in crossing the river of the elephants and reaching the land of 
the Namaquas, a Hottentot tribe of the highest class. Here the 
explorers found natives who had rude copper ornaments twisted 
in tufts of their hair, and wore rings of copper and ivory on 
their arms. They entertained the white visitors with cheering 
hospitality and gave a grand dance in honor of the embassy. 
This was the nearest approach to the civilization of the tra- 
ditional empire that had hitherto been reached by Dutch ex- 
ploration, and the return of the adventurers on March n, 
1 66 1, after forty days' wandering, was warmly welcomed by van 
Riebeeck. 

Before two weeks had passed he had another excursion under 



66 THE DIAMOND MINES OF SOUTH AFRICA 

way led by Corporal Meerhoff, which penetrated into Namaqua- 
land farther than any white man had ever gone, but brought 
back bitterly discouraging reports. It was learned that the 
Namaquas had uncovered some veins of copper and iron ore 
and had some crude process of smelting and working both 
metals, but it did not appear to be practicable to undertake to 
open mines at points so far from the Cape in a region that for 
many months in the year was a torrid desert. There was no 
trace of gold or rumor even of any distant land of gold. Over 
every day's march was the hanging terror of death by thirst or 
hunger or savage attack. 

Still the unflagging commandant would not give up the 
search, and in the following November Corporal MeerhofF 
went back with another party of volunteers to Namaqualand, as 
second in command under Sergeant Pieter Everaert. This 
expedition was better equipped for exploration than any previ- 
ous one that had set out from the Cape, and it was three months 
before it returned to Fort Hope. Yet it had nothing new to 
tell only to repeat the same dreary story of painful tramps 
over sun-scorched sands and jagged ridges of rock, of blinding 
whirls of dust and the blare and clash and drench of terrific 
thunder-storms, of sleep broken by nightly alarms, of lurking 
Bushmen and prowling lions. One of the party had been gored 
and trampled to pulp by an elephant, and his comrades counted 
themselves lucky in reaching the Cape fort empty-handed, 
gaunt, and footsore. 

Even after this sickening rebuff, the next year saw a renewal 
of the attempt to reach the elusive empire of Monomotapa. 
Then Sergeant Jonas de la Guerre set out with a little troop of 
adventurers not yet disheartened. But they were not able to 
push their search into Namaqualand as far as former explorers 
had gone, for they could not find a mouthful of water in the 
desert sands, and were in imminent peril of dying from thirst. 
This repulse was a crushing blow to the stubborn spirit that had 
borne so many buffets. The enterprising van Riebeeck had 
been transferred to the government of Java in the previous year, 



IN TRADITIONAL OPHIR LAND 67 

and his successor was a man of much fainter heart and energy. 
So for nearly a score of years the search for the traditional 
empire lagged, although there was a considerable show of less 
venturesome prospecting. One notable undertaking was the 
despatch of a party of expert assayers and miners from the 
Netherlands to Cape Town in 1669 by tne Dutch East India 
Company, with instructions to search for any promising outcrops 
of ore in the region of the Cape. This party prospected for 
several years, but found nothing to inspire any investment in 
mining. 1 

A revival of the dazzling old visions came in 1681, with the 
appearance at the Cape of a party of Namaquas bearing pieces 
of rich copper ore. This exhibit spurred the East India Com- 
pany to direct another exploration of Namaqualand. Then the 
commandant at the Cape was a man of the stamp of van Rie- 
beeck, commander Simon van der Stel. He was quick to 
despatch a company of thirty soldiers, a draughtsman, and a 
reporter to make the venture so often tried in vain. Again, after 
months of struggle, the desert drove them back. Van der Stel 
then resolved to make an effort far surpassing any put forth 
before by adventurers from the Cape. He formed a party of 
forty-two white men, soldiers, miners, and draughtsmen, with ten 
Hottentot servants and guides. The expedition was provisioned 
for four months, and equipped with two boats, a train of wagons, 
several horses, and a herd of pack oxen. Ensign Olaf Bergh 
was put in command and led his company on to Namaqualand. 
But it was the same old story. No strength of men or oxen 
availed against the desert. No rain had fallen in the wilderness 
north of the Olifants River for twelve months, and the whole 
region was an arid waste without a trickle of moisture. So 
Bergh and his companions faced about in despair, and marched 
back to report their failure. Sergeant Izaak Schuyver and 
another forlorn-hope party tried their luck in the following 
year, and pushed over the desert a little farther than Bergh, but 
brought nothing back except a sack of copper ore on a pack ox. 
1 " South Africa," Theal, Vols. I and 2. 



68 THE DIAMOND MINES OF SOUTH AFRICA 

As a last resort the unflinching commander van der Stel 
resolved to head an exploring party himself. He obtained 
special permission from the directors of the East India Com- 
pany, and his expedition was ordered in keeping with his distinc- 
tion as the head of the Dutch power at the Cape, and with the 
labors and perils of the venture. He left the Castle of Good 
Hope, August 25, 1685, with fifty-six white followers and a 
troop of Hottentot attendants. Twenty-three wagons and carts 
were packed with supplies. Besides the draught teams, there 
were two hundred spare oxen, thirteen horses, and eight mules. 
For the dignity and comfort of the commander there was a 
coach, but this touch of parade was chiefly introduced to impress 
the native tribes and possibly a negro emperor with the grandeur 
of the sovereignty despatching such an embassy. 

The time of year chosen for the start was precisely the same 
as that picked for the expedition of Bergh two years before, but 
the difference in the face of the country would amaze any one 
who had never seen the magic of rain-falls on South African 
deserts. Fresh, juicy grass and vernal flowers were sprouting 
from a soil of seemingly lifeless sand. Birds were building 
their nests in the leafy thickets, insects were creeping or buzzing 
in swarms, and a myriad of butterflies were fluttering their gay 
wings over the green sward and blossoms. After years of 
drought there had come a season of heavy rains. The arid 
sands were soaked, torrents foamed through the windings of 
the dry water-courses, and the region north of Olifants River, 
which had been an impassable barrier to so many explorers, was 
quite easily penetrated by the cumbrous procession of van der 
Stel. Van der Stel's farm and residence were near the present 
town of Somerset West and not far from Stellenbosch, which 
was named after him. His fine old house, " Vergelegen," is 
still one of the remarkable landmarks of these sturdy old Dutch 
settlers. They planted avenues of oaks, camphor trees, and 
pines, which to-day tend to make Cape Town and its environs 
one of the most charming spots on the face of the earth. The 
old picture of van der Stel's house, " Vergelegen," shows it 



IN TRADITIONAL OPHIR LAND 



69 



partly hidden by a huge camphor tree, which measures nine feet 
in diameter. 

As the expedition advanced, it found various promising 
showings of copper ore, and the croppings were particularly rich 
in a range lying a little below the meridian of 30 S., where one 
peak was singled out as " copper mountain." Van der Stel had 
succeeded in reaching the line of the supposed location of the 
golden city of Vigiti Magna, and he pushed his search along 




Vergelegen. 

this line to the Atlantic, but he could nowhere pick up a trace 
of the traditional city or any other vestige of the realm of 
Monomotapa. He did not even meet with any strange mon- 
sters or romantic adventures, except perhaps the charge of a 
huge rhinoceros, which upset his coach and forced him to fly for 
his life. After six months of travel his notable exploring party 
came back to the Cape, without any tidings of good cheer to 
the founders of the colony. The only relic of the tradition of 
empire left in the lands it had traversed was the attaching of the 
name of Vigiti Magna to the great river first shown on any map 
in the chart of this exploration. It had found rich copper ore 
in Namaqualand, but the deposits were too far from the base 



70 THE DIAMOND MINES OF SOUTH AFRICA 

of transportation and supply to warrant the undertaking of 
mining. 1 

Van der Stel was fitly rewarded, four years later, by an ap- 
pointment as the first governor of the Cape Colony, in recog- 
nition of his exploring enterprise and other displays of energy ; 
but his pricking of the painted bubble of Vigiti Magna was a 
bitter disappointment to the Dutch East India Company, and a 
grievous thing to all adventurers filled with the conceit of a cen- 
tury of tradition. It was true that Davaque or some other glit- 
tering city might lie farther to the east and north than any point 
yet reached by Dutch explorers, but with the growing familiarity 
with the land and natives of southern Africa there was a swelling 
discredit of the fine tales of the Dutch and Portuguese roman- 
cers. The myth of the realm of Monomotapa was practically 
starved to death at the close of the seventeenth century, and 
unfortunately the greatly persistent daring of the Dutch explor- 
ers grew cold with its impulse. When adventurers began to 
disbelieve in the marvellous empire and even doubt the location 
of the mines of Solomon and the throne of Sheba, there was no 

f 

very potent lure in the dusty karoos and rocky ravines of South 
Africa. No discovery of ore, except possibly of the precious 
metals, was likely to be of any reward to a prospector, and it 
was even questionable whether rich veins of gold or silver could 
be successfully opened and worked at any considerable dis- 
tance beyond the narrow range of the Dutch settlement at the 
Cape. 

So the credulous search for Ophir and the mythical realms in 
Africa came to an end, and for more than one hundred and fifty 
years there was little life in the tradition of King Solomon's 
mines, until its embers were rekindled by the daring advances 
and glowing fancies of the intrepid explorer, Karl Mauch. In 
1858 Mauch marked the Lydenburg district as a probable gold- 

1 "South Africa," George McCall Theal, Vol. I, pp. 370-380. 

These copper mines came into possession of an English company known as 
the Cape Copper Company in 1853, since which time copper to the value of 
^11,000,000 has been produced. 



IN TRADITIONAL OPHIR LAND 71 

field, and in 1871 he won the honor of reaching and first clearly 
describing the extraordinary ruins of Zimbabwe and its adjacent 
gold-fields. Unfortunately for his credit as an archaeologist he 
insisted on the fancy that the old building on the hill was a copy 
of King Solomon's temple on Mount Moriah and that the lower 
ruins reproduced the palace inhabited by the Queen of Sheba 
during her stay of several years in Jerusalem. 1 This does not 
impair, however, the probable accuracy of his main contention 




that he had revealed part of the ancient workings of the people 
who furnished the flow of gold to Arabia and Judaea in the days 
of King Solomon. 2 

^'The Ruined Cities of Mashonaland," J. Theodore Bent, London, 
1896. 

2 " It was really (Adam) Renders who first discovered these ruins three 
years before Mauch saw them, though Mauch and Baines first published them to 
the world, and they only described what the old Portuguese writers talked of 
hundreds of years ago." E. A. Maund, " Geo. Proc.," February, 1891, 
p. 105. 



72 THE DIAMOND MINES OF SOUTH AFRICA 




Entrance to Boschendal. 



The extent of these old workings has been proved beyond 
doubt by the reports of Hartley, Mauch, Baines, Nelson, and 
later explorers, and a precise and graphic study of Zimbabwe and 

other ancient structures 
in Mashonaland was 
made in 189192 by J. 
Theodore Bent and his 
associates in the expedi- 
tion chiefly promoted by 
the Royal Geographical 
Society, the British Char- 
tered Company of South 
Africa, and the British 
Association for the Ad- 
vancement of Science. Bent's expedition located Zimbabwe in 
latitude 20 16' 30" south, longitude 31 7' 30" east; slightly 
differing from the position given by Mauch. 1 Bent holds that 
Zimbabwe is of Abantu origin and may be freejy translated 
" Here is the great kraal," 
meaning the kraal of the 
native head chief of the dis- 
trict. This name, however, 
marked only the native occu- 
pation of the buildings, and 
Bent sees in the ancient ruins 
and workings " evidence of a 
cult known to Arabia and 
Phoenicia alike, temples built 
on accurate mathematical prin- 
ciples, containing kindred 
objects of art, methods of 
producing gold known to Boschendal. 

have been employed in the ancient world, and evidence of a 
vast population devoted to the mining of gold." 

1 " List of Stations in Mashonaland astronomically observed, with Altitudes," 
by Robert M. W. Swan. 




IN TRADITIONAL OPHIR LAND 



73 




74 THE DIAMOND MINES OF SOUTH AFRICA 




Lekkerwijn. 



Without entering into the varied researches supporting the 
views of Schlechter, Keane, and other leading authorities, it may 
be observed that the main conclusions pithily summarized by 

Professor Keane are strongly 
backed. Ophir was not a 
source of gold, but its dis- 
tributer, as the port on the 
south coast of Arabia through 
which the flow of gold came 
by sea. It is identified with 
the Moschaor Portus Nobilis 
of the Greek and Roman 
geographers. 

Havilah was the land 
whence came the gold of 
Ophir, a great tract in 
southeastern Africa, lying 
north of the Limpopo and largely identified with the range of 
the modern Rhodesia. The ancient gold workings of this region 
were first opened by 
the South Arabian 
Himyarites, who 
were followed (but 
not before the time 
of Solomon) by the 
Phoenicians, and 
these very much later 
by the Moslem 
Arabs. Tharshish 
was the outlet for 
the precious metals 
and stones of Havi- 
lah, and stood probably on the present site of Sofala. The 
Queen of Sheba came by land and not over the seas to the 
court of Solomon. Her kingdom was Yemen, the Arabia Felix 
of the ancients. 




Lekkerwijn. (Back view.) 



IN TRADITIONAL OPHIR LAND 



75 




Bien Donne, Groot Constantia. 

In a word, the "Gold of Ophir" came from Havilah 
(Rhodesia), and was worked and brought thence first by the 
Himyarites (Sabseans and 
Minaeans), later by the Phoe- 
nicians, the chief ports engaged 
in the traffic being Ezion-geber 
in the Red Sea, Tharshish in 
Havilah, and midway between 
the two, Ophir in South 
Arabia. 1 

For sixty years from the 
opening of the eighteenth cen- 
tury there was no considerable 
exploration, or even prospect- 
ing of any consequence, in the 
region north of the meridian 
passing through the Olifants 
River. Yet even in this ap- 




1 A. H. Keane. 



Overmantel and Old Dutch Relics. (Lekkerwijn.) 



76 THE DIAMOND MINES OF SOUTH AFRICA 




Bien Donn, Groot Constantia. 

parent cessation of enterprise there was a continuous progress, 
almost essential to the successful advance of later exploration. 
The Dutch settlement at the Cape was expanding. Year after 
year pioneer settlers pushed out farther from the Castle, moving 

up the river valleys, and cling- 
ing at first to the base of hill 
ranges where the essential sup- 
ply of water was most surely 
attainable. After the taking up 
of the choice locations, later 
comers passed on over the open 
veld, and it was seen that there 
were large tracts of land, un- 
suited to agriculture, which 
would serve well as ranges for 
cattle and sheep. 

For many years, however, 
the raising of wheat was of prime 
Donne, Groot Constantia. importance in the eyes of the 




IN TRADITIONAL OPHIR LAND 



77 




Dutch farmers ; for this product fetched the highest price rela- 
tively, and any surplus was eagerly called for by ships that 

touched at the Cape or by the 

demand for the supply of East 

Indian settlements. In 1685 

the first export of grain was 

shipped, and strenuous efforts 

were made to extend the area 

of land in cultivation. A bo- 
tanic garden had been one of 

the early undertakings of the 

company, to serve as a nursery 

for European, East Indian, and 

native plants, and under the 

direction of Commander van der 

Stel this nursery was made the 

pride of the Cape as an exhibit 

as well as a very serviceable 

source of supply of seeds and 

plants for the garden and farm lands. The growth of the olive 

had been particularly urged, and it seemed at first to be likely 

to flourish, but the 
success of the grove 
of van Riebeeck was 
not attained by plant- 
ers generally. There 
was a considerable 
advance in vine plant- 
ing and the produc- 
tion of wine, and in 
1672 the distillation 
of brandy was begun. 

Doorway, Palmeit Vallei. J { wag hoped that the 

Cape wine could be made an export of consequence, but the taste 
of the Dutch planters preferred a sweet, strong fermentation to 
clear, light wines, and they lacked the skill or the strong desire 



Farm House, Klein Drakenstein. 




78 THE DIAMOND MINES OF SOUTH AFRICA 



to modify their product to compete with French vine growers. 1 
So the only considerable consumption of Cape wine, outside of 

, , the colony, was 

from the crews of 
visiting vessels. 

There was no 
lagging on the 
part of the East 
India Company in 
efforts to stimu- 
late the industries 
of their colony. 
Upon the revoca- 
tion of the edict 

A Wine Farm at Klein Drakenstein. of Nantes (Qct. 

28, 1685) by Louis XIV., the steadfast Huguenots were forced 
to seek new homes in foreign lands, and many were cordially 
encouraged and aided to pass over sea to the young Cape Colony. 





Muller's Farm, Achter Paarl. 



Their expert knowledge of the growth of the vine and olive was 

highly valued, and it was also desired to bring in tanners, har- 

1 " On Veld and Farm," Frances MacNab. 



IN TRADITIONAL OPHIR LAND 



79 




Dutch Farm House. 



ness makers, wheelwrights, metal workers, and other artisans of 
essential service to the spreading settlements of farmers. In 
the allotments of land special care was taken to distribute 
the influx of foreign blood 
so that it must necessarily 
fuse with the main body of 
settlers. This design was so 
well carried out that in a few 
generations the only abso- 
lutely distinct survival of 
this Huguenot migration was 
the perpetuation of the old 
French family names. But 
the combination of these two 
strong strains of blood made 
a compound of remarkable 
character. 

Besides this promoted Muller . s rarm> Achter Paarl . 




8o 



THE DIAMOND MINES OF SOUTH AFRICA 



immigration of men 
there was an equally 
shrewd effort on the 
part of the company to 
advance the breeding 
of horses, cattle, and 
sheep. Stallions were 
imported from Persia 
to improve the stock, 
which had been falling 
off in size and quality 
though increasing in 
uumber. Spanish rams 
were used to lay the 
foundation of the South 
African breed of meri- 
nos, and the Angora 
goats bore transplacing 
excellently, and soon 

browsed greedily on the coarse grasses of the Cape. 

By the advances of the voortrekkers or pioneer farmers the 

range of settlement was extended so far in 1761 that the start of 




Palmeit Vallei, Klein Drakenstein. 



' V 




Muller's farm, Achter Paarl. 



IN TRADITIONAL OPHIR LAND 



81 




Mooi Kelder, Lower Paarl. 

the first large exploring party since the return of the van der 
Stel expedition was made in that year from a rendezvous near 
the mouth of Olifants River. This party was led by Captain 
Hendrik Hop of the burgher militia, and was made up of seven- 
teen whites and 
sixty-eight half- 
breed Hottentot 
servants. It 
started in August 
and advanced on 
the track of the 
former expedi- 
tion, passing the 
Copper Moun- 
tains of Little Plaisis de Merle, Groote Drakenstein. 

Namaqualand, and reaching the river Vigiti Magna on Septem- 
ber 29. This river was familiarly called by the colonists the 
Groote (Great) River, and held this name until both the tradi- 




82 



THE DIAMOND MINES OF SOUTH AFRICA 




tional and common names were supplanted by a new christening 
in 1779, when Colonel Robert Jacob Gordon, commanding the 

garrison at Cape Castle, led 
another expedition up the 
river, and named it Orange 
in honor of the stadtholder. 

Hop's exploring party met 
a troup of giraffes soon after 
crossing the Groote River, 
and won the distinction of 
furnishing the first skin of a 
giraffe from South Africa to 
the Museum of the Univer- 
sity of Leyden. But except- 
ing this novel chase there was 
little to attract the explorers. 
The sun scorched them relent- 

Donkerhoek, Groot Drakenstein. i i i i i 

lessly in the open desert, and 

they could nowhere find water except in the deep sand-pits dug 
by the roving natives. Sometimes there was a shallow puddle 
at the bottom of 
one of these pits, 
and even when the 
sand was barely 
moist, further dig- 
ging to the under- 
lying stone would 
sometimes yield a 
trickle of water. 
Still they pushed 
on stubbornly to 
the farthest point 

Vet reached from AWineCellar. H erd of Cape Goats. 

the Cape, in latitude 26 18' S., before turning back to bring 
home their discouraging story. 

It was thirty years before this advance was outstripped by 




IN TRADITIONAL OPHIR LAND 



Willem van Reenen, of the farm Zeekoevlei on the Olifants 
River. This adventurous farmer set out in 1791 with four 
fellow colonists and a number of Hottentot servants, and reached 
on the 1 8th of November the end of the trek of Captain 
Hop's party. Prowling Bushmen and lions beset their camps 
continually, and in January, 1799, tne 7 na d to beat off a fierce 
swoop of a party 
of Namaquas. Yet 
they pressed on 
until March 14, 
when they came 
to a little oasis 
which they named 
Modder Fontein, 
or muddy spring. 
Then they turned 
back after a few 
days' rest, and 
plodded home to 
the farm Zeekoe- 
vlei, which they 
reached on the 
2oth of June. 
They had killed 
sixty-five rhinoc- 
eros and six 
giraffes, without 
reckoning their 
bag of smaller game, and brought back exultantly wagon loads 
of copper ore, which they supposed to be gold until their hopes 
were blighted by assayers at the Cape. 1 

The depressing reports from these expeditions were not the 

least of the straws that finally broke the back of the Dutch East 

India Company. For nearly a century and a half their colony 

in South Africa had been a continual drain and burden. All 

1 South Africa," George McCall Theal. 




Tatr, 1757. 



84 THE DIAMOND MINES OF SOUTH AFRICA 




An Old Farm House, Lower Paarl. 



the expedients and efforts of the energetic directors of the com- 
pany in the seventeenth century, and such faithful servants as 
van Riebeeck and van der Stel, had failed to develop any mines 
or any product for export of any considerable importance. With 

the beginning of the 
eighteenth century there 
was an evident drooping 
in the enterprise of the 
company, and a drift 
toward hopeless discour- 
agement, which culmi- 
nated in 1794 with the 
declaration of bankruptcy. 
The company's debt was 
^10,000,000 sterling, 
and its credit was utterly exhausted. It could no longer under- 
take even to maintain a feeble garrison at the Castle for the 
defence of its colony. Issues of depreciated and irredeemable 
paper had driven out all gold and silver from circulation at 
the Cape. Debts could be paid in this paper, which was legal 
tender, but nobody would receive it in exchange for goods 
except at such a discount that there was a general resort to 
barter. Internal trade was para- 
lyzed, and a little wheat, wine, 
and tallow was all that could be 
squeezed out of the colony for 
export to Java and India. The 
straggling settlers on the north- 
ern frontier were continually 

fighting with the Ishmaelite Farm House. Achter Paarl. 

Bushmen, and the Kafirs on the northeast were still more 
harassing and formidable. Every kraal was a rude fort and 
every family a garrison. Ammunition was growing scarce and 
costly, and there was no hope of succor from the Castle at the 
Cape. 

In view of this patent collapse, the stretching out of the 




IN TRADITIONAL OPHIR LAND 




strong arm of Great Britain to seize the Cape in 1795 should 
have been as welcome as rescue to a wreck. Then for the first 
time a power took hold of the way station of East Indian trade, 
and its straggling offshoots, that had the strength and the skill 
and the far-reaching conception to 
do more than repress savage on- 
slaughts and defend grazing 
grounds, to open great mines, 
to convert arid karoos into irri- 
gated plantations, to extend the 
network of railways, and stretch 
in time the steel band of civiliza- 
tion across the darkest zone of 
Africa. This Britannia has done 
and is doing, either in her imperial 
way, or by the hands of the sons 
who have labored to make her 
greater. 

But the coming of this saving 
and transforming power had the appearance, at the time, of a 
hostile attack. The Netherlands, in 1793, were wholly under 
the thumb of the new French republic, and war was declared 
against Great Britain through controlling French influence. 
There had been some revolting against the further collection 

of taxes by officers of 
the East India Com- 
pany, but the colonists 
as a body did not want 
any foreign interference. 
So the little garrison in 
the Castle at the Cape 
put on a defiant front, 

Cape Cart. anc j ra HJ e d tO itS Support 

a number of burgher volunteers when a strong British fleet sailed 
into Table Bay in the first week of September, 1795. It was ap- 
parent, however, even to the boldest Dutch defender, that resist- 



Brand Solder (Fire Loft). For the 
prevention of fire. 




86 THE DIAMOND MINES OF SOUTH AFRICA 



ance was hopeless, and Cape Town, with its castle and garrison, 
surrendered to Admiral Sir George Elphinstone and General Sir 
Alured Clarke, on the sixteenth of September. So was ended 
one hundred and forty-three years of rule of the Dutch East 
India Company, and from this date British ascendancy in South 
Africa began. There was a brief intermission, it is 
true, some years later, when the treaty of Amiens 

(1802) transferred the 
Colony to the Ba- 
tavian Republic. 
But the breaking 
out of war again 
in the following 
year ruptured the 
treaty, and ex- 
posed the Cape 
Colony again to the hazard of capture, which actually followed 
early in January, 1806, when Cape Town was retaken by Major 
General David Baird. From that time the Cape was held con- 
tinuously by the strong arm until the convention at London, 
August 13, 1814, when all claims of the Netherlands to South 
Africa were extinguished by cession, and Great Britain became 
the heir of all the Dutch advances from the Cape of Good 
Hope. 1 

1 "South Africa," George McCall Theal. " Precis of the Archives of the 
Cape of Good Hope," H. C. V. Leibrandt. " South Africa," Augustus Henry 
Keane. "Heroes of South African Discovery," N. D'Anvers (Henry Bell). 




: 







CHAPTER III 



THE PIONEER ADVANCE 




HEN Lord Charles Somerset came to the Cape 
as the first Governor of the Colony after the 
cession, how slight and infirm was the hold of 
any civilization on the indurated barbarism 
of the vast expanse of Africa south of the 
equator! In the three hundred years that had 
passed since Vasco Da Gama made known the bounds of the 
continent, the outer rim of the traditional Ophir land had barely 
been pierced. From the Atlantic side the Portuguese had not 
pushed beyond a fringe of trading posts on the Lower Guinea 
coast, and were clinging feebly to insignificant stations along the 
shores of the Mozambique channel. The Dutch grip was more 
obstinate, in spite of all disappointments, but the range of their 
advance was only a few hundred miles from the Cape, and out- 
side of Cape Town the population was a mere sprinkling on the 
face of the land. When the British first wrested the Cape from 
the Dutch, Earl Macartney, who held the government in 1797, 
defined by proclamation the bounds of the Colony. It only ran 
east to the Great Fish River and on the north to the Zuurberg 
Mountains and the southern edge of Bushman's land, trending 
up to the Kamiesberg, and thence along the coast to Buffels 
River in Little Namaqualand. The total extent was roughly 
120,000 square miles, merely the extreme tip of South Africa, 
and the entire population, both white and black, was reported 
to be less than 62,000, or about one person to every two square 
miles. This was a petty fringe on the skirt of the dark continent. 
Not only was the Colony weak in numbers, but it was seem- 
ingly without any uplifting leaven of enterprise and ambition. 

87 



88 THE DIAMOND MINES OF SOUTH AFRICA 

For generations the Dutch settler had been treading in the foot- 
steps of his forefathers without any wish to stride ahead. What 
they had done, he would do if he could. No new way of work- 
ing or living or thinking was as good to his mind as the old way. 
The pioneer farmer and grazier had often been constrained to 
pack all his goods on the backs of oxen or in a wagon with his 
wife and children. A little hut of " wattle and daub " sheltered 
the family. Rude frames of wood overlaid with raw hide strips 
were their bedsteads, and sheepskins, their bedclothes. They 
cooked their food on the coals of wood-fires or boiled it in an 
iron pot. They cut their meat with clasp knives and drank 
from tin cups. A big chest served them for a table. Their 
house floor was the bare earth, unless a strip was covered with 
a wild beast's skin. Their children were brought up from their 
birth in this habit of life and the lack of comforts was not to 
them a privation. Their standard of living was scarcely higher 
than that of the imported Guinea slaves who worked for them, or 
of the native tribes that surrounded them. Their isolation from 
civilized society and their life in the wilderness in familiar con- 
tact with slaves and savages was inevitably degrading. When the 
English took the Colony, there was not a bookstore or a single 
good school in it, and outside of Cape Town almost the only 
tutors were soldiers who were allowed to live with the farmers. 1 
Still there was one sustaining and universal spirit which kept 
even the rudest grazier from sinking to the barbaric level. They 
clung to the God of Israel and to the Bible as God's revelation. 
They never wearied of searching the Scriptures, and they prayed 
with the fervor and faith of the old Covenanters. Their creed 
was the strait and narrow way of Calvinism and the synod of 
Dordrecht, and they turned to the Old Testament as confidingly 
as to the New for guidance. They recognized the holding of 
slaves as a practice permitted to Israel, and they made bond ser- 
vants of the Hottentots in their apprenticeship contracts. In 
their eyes the Bushmen were Ishmaelites and the Kafirs Philis- 

1 "South Africa," George McCall Theal. "Handbook to South Africa," 
S. W. Silver & Co. 




THE PIONEER ADVANCE 89 

tines, who were cumbering the ground that might be occupied by 
God's favored people. 1 But the settlers were phlegmatic and 
peaceful by nature, content with their bare living, and with no 
ardor for extending their bounds by conquest. An extraordinary 
impulse was needed to convert them into adventurers and wan- 
derers in the desert. 

This impulse was given by the capture of the Cape, the influx 
of jostling immigrants from Great Britain, new and vexing legis- 
lation, and disasters to crops which exalted the comparative value 
of pasturage lands. 2 At the opening of the administration of 
Lord Charles Somerset there was a marked effort on the part of 
the Home Government to promote the growth of the Colony. A 
regular mail packet service was established between England and 
the Cape, and ^50,000 were voted by Parliament in 1819 to be 
disbursed in aid of emigration to South Africa. This contribution 
was a powerful stimulus, and it is estimated that nearly 5,000 
new settlers of British birth were added to the population of 
Cape Colony from March, 1820, to May, 1821. 

Unfortunately the South African climate in 1820 and the 
years immediately following was peculiarly aggravating. In 1819 
there had been a heavy wheat crop and the consequent tempta- 
tion to farmers to extend their wheat growing. So they did, but 
the crop of 1820 throughout South Africa was fatally blighted. 
The next year's crop fared no better, and thousands of farmers 
were ruined and brought even to the verge of starvation. Rations 
were distributed by the Colonial Government in the fall of 1821 
to those who had no means to buy food, but the unrelieved 
suffering was widespread. Following hard on this scourge of 
blight came the prodigious floods of October, 1823, when it 
seemed to the colonists in the eastern districts as if the heavens 
were open for another deluge. Rain fell in torrents for days 
without ceasing, and overflowing rivers ran foaming to the sea, 
carrying millions of tons of earth in their turbid floods as well as 
the shattered houses of settlers who had barely time to fly for 

1 "Impressions of South Africa," James Bryce. "South Africa," Theal. 

2 "Annals of Natal," John Bird, p. 505. 



90 THE DIAMOND MINES OF SOUTH AFRICA 

their lives. These staggering rebuffs in the face of the new emi- 
grants were greatly demoralizing. Some fled from the Cape in 
despair, and many more wrote home to their friends that the Col- 
ony was hung between flood and famine, and that the greater part 
of South Africa was a dismal Karooland. Still there was a notably 
plucky rally and an immediate turning to other resources when 
wheat cultivation was shown to be an uncertain reliance. Cattle 
and sheep breeding was largely extended at once, and in 1828 
hides and skins ranked only second to wine in the list of 
exports. 1 

The failures in wheat growing and the resort to pasture land 
were strongly moving influences urging on the advance of pio- 
neer settlers from the southern river valleys north and east over 
the veld into unclaimed territory. This natural flow of migra- 
tion was greatly swelled and impelled by the clashing of the old 
settlers with the newcomers from Great Britain, and by their 
resentment of British control and administration measures. By 
the census of 1819 the white population of the colony was 
42,217, and outside of Cape Town this people was almost 
wholly of Dutch descent or of the fused Dutch and Huguenot 
strains. It was inevitable that a stock of such breeding and tra- 
dition should be impatient of any ordinances or ways except its 
own. It was peculiarly irksome to bow to a nation which had 
captured the Cape by the strong arm, and was only represented 
by a small minority of the settlers. The inevitable heart-burn- 
ing was aggravated by the contact and rivalries of the new and 
old settlers. Neither faction had the knowledge or temper to 
recognize the best traits in the other and show tolerance for dis- 
similar habits and prejudices. The Dutch boer has an old 
Anglo-Saxon root and is simply correspondent to the German 
bauer, a farmer or countryman ; but in the English mouth all 
the Dutch colonists were lumped as Boers, and in the English 
eye Boer was too often confounded with the clownish boor. 
The Boers faced this contempt with a glowing resentment that 
burned like a slow-match. 

1 South Africa," Theal. 



THE PIONEER ADVANCE 91 

In the new measures of government there was a succession 
of vexations also to colonists attached to the old customs and 
ordinances. The expense of the new colonial establishment was 
a grievance. The adjustment of the currency aroused bitter 
complaint. The substitution of English for Dutch in official 
papers, and the abolition of the old Dutch courts, were heavy 
humiliations. But the keenest resentment was excited by the 
measures designed for the protection of Hottentot bond servants 
and free natives, and the emancipation act of 1833. There had 
been a rapid increase in the importation of slaves from Guinea 
after the first conquest of the Colony by the British, but in 1807 
the last cargo of slaves was landed at Cape Town, and the slave 
trade was formally brought to an end by law in the following year. 
Still the colonists continued to hold and breed slaves as their 
fathers had done, and there were 35,745 slaves in the Colony 
when the emancipation act went into effect on the first of Decem- 
ber, 1834. These slaves were valued at _3 ,000,000, but only 
,1,200,000 were appropriated as compensation to their owners. 
The loss fell heavily on many owners already sinking under the 
weight of mortgages, and there were rumblings and outpourings 
of bitter indignation. The deficiency in compensation was called 
Imperial confiscation, and the Boers resented it sorely, not merely 
on the score of the loss measured in money, but as a crowning 
instance of their political subjection. 1 Alien Imperial rule was 
the deep-seated grievance which was the underlying and impel- 
ling cause of the extraordinary exodus from Cape Colony called 
the Great Trek. 2 

In 1835 Louis Triechard led out the first pioneer company 
of this migration, and his advance into the wilderness beyond the 
bounds of the Colony was followed by a succession of slow-mov- 
ing caravans pushing northeast to the head waters of the Orange 
River and the terraces of Natal, and moving on, in course of 
years, across the Vaal to the Limpopo water-shed. This out- 
push of pioneers in large parties, overcoming all barriers of 

1 " Annals of Natal." " South Africa," Theal. " The Great Trek," Henry 
Cloete, her Majesty's High Commissioner for the Colony of Natal. 2 Ibid. 



92 THE DIAMOND MINES OF SOUTH AFRICA 

mountains and deserts, and fearlessly venturing into the strong- 
holds of the fiercest native tribes, undoubtedly hastened and 
secured the acquirement of the marvellous diamond and gold 
fields of South Africa. The march of the caravans and the 
winning of the land was a drama full of barbaric color and 
movement. 

At the time when the Cape first fell into the hands of Great 
Britain, there was an insignificant tribe, the Amazulu, living in 

kraals on the banks of 
the river Unvolosi, which 
flows into the Indian 
Ocean at St. Lucia Bay. 
In their name there was 
an arrogance of high de- 
scent, for its meaning is 
" the people of the sky "; 
but the Amazulu had then 
nothing else to brag of, 
and while their head chief, 
Senzanzakona, lived, there 
was no terror in the Zulu 
name. But there was a 
son born to Senzanzakona 
in or near the year 1783 1 
who made the Amazulus 
masters of a region far 

Zulu Chief Cetawayo and Part of his Family. , . . . r 

exceeding any bounds of 

the Kalangu Monomotapa, and stamped his name across it in 
indelible blood. 2 

The boy was called Tshaka or Chaka, which, in the Sechuana 
tongue, is " battle axe." There is another tracing of his name 
to Cheka, a wasting disease afflicting his mother. In either 
translation the name was ominous. But this chief's son had no 
deformity that an eye could see. When he came to manhood, 
a sculptor would have picked him as a model of his tall, athletic 
1 "South Africa," Theal. "Annals of Natal." 2 Ibid. 




THE PIONEER ADVANCE 



93 




race. He was more than six feet in height, and every inch was 

pulsing with vigor. No rival could leap as high or hurl an 

assagai as far. In later life his 

shapely features were swollen with 

ugly passions and debauch, and 

his lithe body was overlaid with 

fat, but he never lost the beauty of 

his deep-set, brilliant black eyes, 

fringed with their long, curved 

eyelashes. 

For some cause Chaka, while 
only a lad, was forced to fly for 
refuge to Dingiswayo, chief of the 
Abatetwa, the master tribe of 
the district. Under protection 
of this chief he was made a sol- 
dier, and took by craft the head- 
ship of his own Zulu tribe when 
his father died. Then he was Zulu Prince ' Dinizulu ' 

able to betray and put to death his protector Dingiswayo, and 

spread his mastery by force or 
,~t terror over the surrounding 

' ' * ''.''-'-' .. A I ' I 

'/ tribes. As he grew in power he 
showed an unfolding genius for 
war and command. He pressed 
every young and strong man 
within reach into his army. He 
marshalled his men in impis or 
regiments. He discarded the old 
bunch of assagais and armed each 
man with a single, short-handled, 
long-bladed unkonto or spear, 
and protected him with a shield 
of oxhide. He aimed with his 
weapon to make every fight hand to hand, where every man 
must kill or be killed. If a soldier lost his spear he was 




Zulu Family. 



94 THE DIAMOND MINES OF SOUTH AFRICA 

doomed to die, unless he could show another in place of it, torn 
from an enemy. 

No barbaric figure was ever more terrific and martial than 
the Zulu soldier in war-dress. Chaka's hair was cut close, 
except on the top of his head where the thick, crisp locks were 
matted or moulded into a ring made of a tree gum and polished 
to the likeness of ebony. Thick folds of otter pelt were wound 
round his head and great earrings of carved sugar-cane hung 
from the cut lobes of his ears, which were covered with pads of 




A Zulu and his Ten Wives. 

jackal's skin. From this turban projected two feet or more a 
jet-black crane feather, waving with every toss of his head. A 
circlet of twisted monkey and genet skins hung over his breast 
and back, and from his waist a thick flexible kilt of twisted skins 
hung to his knees. Bands of short-cut white oxtails circled his 
legs and arms, and the ruffles round his ankles made his bound- 
ing feet oddly like the winged Mercury. In his right hand he 
grasped his spear and swung at his left side his oval shield of 
white oxhide. Now pin with thorns a dozen bunches of the 
red feathers of the louri in the crisp tufts of his crown and scat- 



THE PIONEER ADVANCE 



95 




Zulu Kraal and Huts. 

ter some other brilliant feathers on a circlet above his breast, 
and see Chaka dressed for parade. 1 

Then fancy the marshalling of an army of men like him, for 
the chieftain in arms was one of ten thousand. When the lead- 
ing division marched on in review, every man was more or less 
closely the image of Chaka. These picked men were his Unbala- 
bale or Invincibles, scarred veterans who had never been beaten. 
They bore white shields marked, like their chief's, with a black 
spot, and behind them followed in grade of honor divisions with 
red-spotted shields, gray shields, and black shields. Only the 
Invincibles had kilts of skins, the others wearing instead a trap- 
ping of oxtails. As these fierce troops marched on before 
Chaka's keen eye, the men of chief mark would bound from the 
ranks and show a marvel of vaulting, darting to and fro, whirl- 
ing of spears and mimicry of fight, in which few athletes could 
compare with the supple Zulu. 

In formation for battle Chaka curved the van of his impis 

1 "Annals of Natal," pp. 90100. 



96 THE DIAMOND MINES OF SOUTH AFRICA 

like a crescent. He called the end his horns and the centre his 
breast. This was the old array of the warring Bantu tribes, but 
Chaka greatly strengthened it by a formation behind in an 
oblong block of men held in reserve to repel any break in the 
crescent or reenforce it when wavering. His force of disciplined 
soldiers ranged up to fifty thousand strong. 




Zulu Hut in course of Construction. 

With this prodigious engine of war shaped to his hand, he 
overran all the country from Delagoa Bay to the Unzimvulu 
River and far into the interior, scourging its face mercilessly. 
Some of the terrified tribes in his way were blotted out com- 
pletely. " There was a white mark from the Tugela to Thaba 
N'chu, and that was our bones," said an old Hlubi to Theal, the 
historian of South Africa. Sometimes stragglers escaped to lurk 
in mountain recesses. These wretched survivors of the scourge 
were covered by one new and pitiful name, Amafengu, because 
their first cry to strangers was Fenguza, " we want." Only one 
tribe held Chaka in check, the warlike Amaswazi, which stub- 
bornly guarded their mountain paths and cliffs. Even the fierce 
Amangwane were forced to fly before Chaka's resistless impis ; 
but they kept massed together, and in their retreat drove off or 
massacred most of the tribes between the Orange and the Vaal 
rivers. Then the Amangwane, still hot pressed by the Zulus, 



THE PIONEER ADVANCE 



97 



began to rub against the frontiersmen of Cape Colony. This 
inroad was bravely met by a muster of a thousand soldiers and 
Boers under Lieutenant Colonel Somerset, who finally put the 
Amangwane to utter route in a sharp battle, August 27, 1828, 
near the banks of the Bashil River. 1 

Chaka was a warrior capable of measuring the efficiency of 
the white man's organization and firearms. When the Aman- 
gwane were thrown back, the Zulu chief withdrew his own impis 
without risking a collision with the whites. A few weeks later 
he was murdered by two of his half brothers and his best-trusted 
attendant. Dingaan, his half brother, and one of his assassins, 
grasped the headship of the Zulus, but his succession was dis- 




Zulu Woman grinding Corn. 

puted by the commander of one of the chief divisions of Chaka's 
army, the unruly Matabele. This revolting chief, Umsilikazi, 
was the model of a Zulu warrior, tall, sinewy, shapely, and, 
except in war dress, naked save for a cord around his waist 
from which leopards' tails dangled. A string of little blue 
beads was drawn about his sturdy neck, and three green feathers 
of a paroquet were stuck in his crisp hair. His followers were 
like him, and the wild charge of the legion of such men armed 
1 " South Africa," Theal. Annals of Natal." 



9 8 



THE DIAMOND MINES OF SOUTH AFRICA 



with their keen-bladed spears was a sight that would try the 
nerve of any white soldier. How the rudely armed and undis- 
ciplined Boers would face it was soon to be tested. 

Umsilikazi, revolting from Dingaan, led his Matabele divi- 
sion across the desert to fall upon the country north of the 
Orange River and west of the Drakensberg, the Dragon 
Mountains. Much of this country had been ravaged before 
by the Amangwane, and the Matabele spared nothing that had 
escaped slaughter and pillage. Dingaan sent an army of Zulus 
in 1834 to dislodge his rival, but the warriors of Umsilikazi 




Zulu Women. 



beat back the attack. By the Zulu raids and massacres and 
wars, the whole country from the seaboard of Natal nearly to 
the junction of the Orange and Vaal was desolated, and the 
native tribes of the region almost destroyed. Thus great tracts 
of land were opened to the advance of the migrating Boers, but 
the push of the trekking pioneers soon brought them in conflict 
with Umsilikazi and Dingaan. 

Then the remarkable traits of this peculiar people stood out 
in high relief. To English immigrants, jostling the old settlers, 
the ordinary Boer appeared a Dutch clodhopper, sullen and jeal- 
ous, unkempt in person and dress, immovably set in his traditional 
ways, pig-headed in his obstinate prejudices, a block to every 
suggestion of progress, Pharasaical in his prayers, absurd in his 



THE PIONEER ADVANCE 



99 



customs, and often clutching to the last penny. 1 There were 
some true lines in this partial portraiture, with a natural warping 




Zulus smoking Indian Hemp. 



of prejudice and lack of insight. In face of the foreign intru- 
sion the Boer had something of the instinct of the turtle and 




Old Zulu Women taking Kafir Beer to a Wedding. 

1 "The Great Thirst Land," Parker Gillmore. "South Africa," George 
McCall Theal. "South Africa; a Sketch Book of Men, Manners, and Facts," 
James Stanley Little. 



ioo THE DIAMOND MINES OF SOUTH AFRICA 




porcupine. But in the heart of the wilderness, in his venture- 
some trek over the pathless veld, and in the traverse of moun- 
tains and deserts, he showed what scornful eyes had not seen, 
the self-reliance, the fortitude, and the pluck of the true pioneer. 
He packed his wife and children and all his needful supplies 
in a huge, low-bodied wagon under an arched frame covered with 

waterproof canvas. To this stout 
wagon sixteen strong oxen were 
yoked to the chain or rawhide rope 
forming a trektouw. Every ox was 
a helpmate. Every one knew his 
name and place and resented a 
change in yoking. The Boer and 
his Hottentot helpers spoke to them 
all familiarly, and could cut at will 
a fly from the ear of any one with 
a flick of their long-lashed whip. 
When these prairie-schooners lum- 
bered ofF, creaking and swaying, 
with a chorus of Dutch and native calls, the Boers and their 
sons rode beside them on ungainly flea-bitten horses, trained to 
herding and hunting, and often possessing uncommon bottom 
and speed. 

The Boer was by nature prudent and wary. For comfort 
and safeguard the advance of the Great Trek was in companies, 
camping at night on plain and hillside, with wagons ranged to 
form a rough palisade and kraal. No morning or nightfall ever 
passed without prayers and the reading or recital of Scripture. 
For every step of his way he looked to his God for guidance, 
and he felt that the old promises to the chosen people were 
renewed to him. His faith in the literal inspiration of the Bible 
was unwavering. He did not doubt that the sun stood still at 
the call of Joshua, or wonder at the slaughter of Philistines with 
the jawbone of an ass. In face of every privation and the direst 
peril he was sustained by his certain reliance on the help of One 
who could make a spring gush from the desert rock, or deliver 



Zulu Girls. 



THE PIONEER ADVANCE 



101 



any heathen host into the hands of a few faithful servants. 
But with all this reliant devotion he never forgot " to keep his 
powder dry," and used every opportunity to perfect his skill as 
a marksman. 

Back of his faith and prudence was an unflinching spirit. In 
the uncouth Boer smouldered the fire of an ancestry that charged 
at Ivry and starved at Leyden. Even the women and children 
were dauntless at the 
pinch of need. With 
her white grease-cloth 
wrapped about her face, 
the Boer's vrouw was 
an uncouth object, but 
with her eye on the 
sight of a rifle many a 
fat old woman was a 
guard to be feared. 

No impediments 
nor dangers stayed the 
advance of these pio- 
neers. When a heavy 
wheel dropped into a 
deep gully or earth- 
crack or ant-bear hole, 
it was pried out with un- 
tiring patience. When 
thunder-storms changed 
the red soil to beds of 
mire and the wheels were clogged masses of mud from nave to 
felloe, the mud was laboriously scraped away and the wagons 
tugged to firmer ground. When the violent wrenches and strains 
snapped trektouws and wagon-poles and king-bolts like pack- 
thread, the same inflexible temper relinked the broken touws with 
riems of rawhide, chopped out new wagon-poles, and forged new 
fastenings with rude blacksmith's art. No karoo was so forbid- 
ding and no stream so swollen as to bar the onward march. 




Native Laborers in War Dress. 



102 THE DIAMOND MINES OF SOUTH AFRICA 

The tired Boer snored serenely at night behind the bulwark 
of his wagons, regardless of the wild beasts prowling and sniff- 
ing outside. The giggling calls of the gray and brown jackals, 
the doleful howl of the slinking hyena, even the deep breathing 
sough of the lurking lion, did not open his eyes, and it must be 
a fiercely menacing roar indeed that would lift his head. His 
only haunting dread was the crippling of his march by the 
deadly tsetse fly or the wasting diseases that made his horses 
and oxen the prey of the vulture. 




Trekbok (Springbok) Hunting. 

In the passage of these pioneers the destruction of wild ani- 
mals of all kinds was enormous, partly for the sake of needful 
food, and partly for the skins, but much wantonly and waste- 
fully, for the Boer would rarely let pass a living mark for his 
rifle. Of lesser game there was no attempt to keep tally, but by 
a common report thousands of lions were shot in the march to 
the Transvaal. Any such reckoning must be largely guesswork, 
though there is no doubt that few beasts within range escaped with- 
out the sting of a bullet. But a foe more formidable than any 
multitude of lions sought to bar the progress of the Great Trek. 

The revolting Umsilikazi was the first of the great Zulu 
chiefs to try the temper and the arms of these pioneers. One 



THE PIONEER ADVANCE 



103 



of the larger divisions of the Great Trek, led by Hendrik Pot- 
gieter and Gert Maritz, left the Cape Colony in August, 
1836, and pushed north of the Caledon River. 1 Some of the 
pioneers in this advance were cut off suddenly and killed by 
Umsilikazi. Flushed with this bloodshed, he made a swoop 
with six thousand men upon a part of Potgieter's trek a com- 
pany of a few score men, women, and children. But the startled 
Boers were now on their 
guard. They ranged their 
big, white-tented wagons in a 
square, lashing the wheels to- 
gether with rawhide riems, 
and filling in the chinks in 
their barricade with thorny 
mimosa bushes. In the cen- 
tre of this laager a few wagons 
were placed as a cover for the 
women and children. 

Upon sight of the ad- 
vancing Matabele, all knelt 
and prayed. Then some of 
the men rode out boldly to 
meet the attack with their 
heavy rifles. Their fire was 
deadly, killing, at times, two 
or three at a shot, when their 
guns were loaded with slugs, 
but the impis pressed on, 
driving the Boers back to their laager in a sullen retreat, turning 
to fire as fast as they could reload. Within the laager all was 
made ready for a defence to the death. Back of every wagon a 
little heap of powder and bullets was put on the ground, and the 
women stood by to hand spare guns and reload. It was sternly 
ordered that there should be no shrieking or crying by women 
or children. In silence the rush of the Matabele was awaited. 

1 The Caledon River divides Basutoland from the Orange River Colony. 




Zulu in War Dress. 



104 THE DIAMOND MINES OF SOUTH AFRICA 

On came the impis in raging masses that dashed on every 
side of the laager like surf on a reef, wrenching at the wheels, 
clambering over the canvas, plunging through the thorns. The 
heavy wagons were shaken and swayed, but the lashed barricade 
held fast. The grim Boers met the shock with withering 
volleys, piling up the blacks in bloody heaps around the laager. 
Crouching behind the firing line, the women moulded bullets 
and helped to reload. 

The firing was so deadly and the laager so impenetrable that 
the surges massed against it recoiled. But, after a moment of 
rallying,on came the billows of men, flinging their assagais, and 
howling like madmen as they crashed against the barrier which 
shielded the Boers. They stabbed and slashed at the canvas 
covers in frenzied efforts to cut their way over the wagons, and 
wriggled through the crevices packed with thorn bushes, until 
some, torn, bloody, and gasping, squirmed into the square, where 
the Boer women killed them with knives and hatchets. The 
Boers fired as fast as they could lift their rifles, not stopping to 
use their ramrods, but grabbing handfuls of powder to charge 
their guns, and dropping in slugs with scarcely any wadding. 

So intense was the strain of that hour that even these men 
of iron nerve were entranced. " Of that fight," wrote one, 
" nothing remains in my memory except shouting and tumult 
and lamentation, and a dense smoke that rose straight as a plumb 
line upwards from the ground." 1 

Four times the black impis charged and four times their 
onset was beaten back before Umsilikazi drew off his men. 
The field around the laager was a fearful sight, and the white 
tops of the barricade were slashed into strips and dripping with 
blood. Seventy-two stabs were counted in the cover of one 
wagon, and eleven hundred and seventy-two assagais were flung 
through into the camp. But none of the stout defenders were 
killed, and all joined devoutly in a psalm of thanksgiving. 

In retaliation for this attack Hendrik Potgieter and Pieter 
Uys led a troop of one hundred and thirty-seven in a swift 

1 "Annals of Natal," p. 375. 



THE PIONEER ADVANCE 



105 



march and onslaught upon the main division of Umsilikazi. 
The attack was so well timed and aimed that the array of fierce 
impis was shattered and their chief was driven in flight to the 
wilderness beyond the Limpopo. There, in the present Mata- 
beleland, Umsilikazi brought together the remnants of his 
people, and ruled in awe of the pioneers until his death in 1870. 

Hard upon the defeat of 
Umsilikazi came the greater 
clash with Dingaan, when the 
trekking Boers crossed the Dra- 
kensberg or Dragon Mountains 
to the terraces of Natal. This 
cunning and tricky chief made 
smooth professions of friendship 
to the Boers at first. He wel- 
comed as allies the company 
headed by Pieter Retief and re- 
ceived the commander at his 
kraal. The chief's house was a 
spherical hut about twenty feet 
in diameter. Its floor was pol- zuiu-jim cameei. 

ished till it shone like a mirror, and its roof was supported by 
twenty-two pillars of wood completely covered with beads. 
Around this house were seventeen hundred ruder huts which 
Dingaan used as barracks for his impis, and each hut would 
cover twenty men. 

After some parleying Dingaan signed a cession of the greater 
part of the present territory of Natal to the Boers. To cele- 
brate the compact he invited Retief to visit him again with his 
companions. It was agreed as an exhibit of good faith that no 
arms should be taken into the chiefs kraal. So Retief and 
some sixty other Boers, with forty Hottentot attendants, piled 
their arms outside the kraal, and came in before Dingaan, who 
was sitting in an arm-chair in front of his hut. Two of his 
impis were formed in a circle about him. The Boers took their 
seats on the ground within the circle, and cups of utywala or 




io6 THE DIAMOND MINES OF SOUTH AFRICA 

native beer were offered them to drink. But when they put 
their lips to the cup, Dingaan cried out, " bulala amatagati," 
" kill the wizards." At this cry his Zulus fell on their helpless 
guests in overwhelming mass. A few Boers had clasp-knives, 
and the others met the rush with naked hands, but all were 
overpowered in a moment and dragged over the ground to a 
hill near by, called Hloma Mabuto, or the mustering of the 
soldiers. Here their heads were crushed with knob kerries, 
and their bodies were flung into heaps. Retief was forced to see 
the horrid murder of all of his companions. Then his heart 
and liver were cut out and taken to Dingaan, and the mutilated 
corpse was cast on the heap of dead. 1 

None of the Boers in the trap escaped, and after the mas- 
sacre the Zulus poured out to raid the scattered camps of the 
pioneers. They were finally beaten back at Bushman's River, 
after they had killed many trekkers and carried off their cattle, 
and the mounted Boers followed their retreat for days. But 
the Zulus were quick to turn and strike again like fierce hawks, 
and within two months they swooped down upon the English 
settlers and native blacks of Natal and cut them off almost to a 
man. 

The trekking Boers were hard pressed. Pieter Uys was 
killed in ambuscade, with his son, a boy of fourteen, and a num- 
ber of his men. When Uys was fatally wounded, he urged his 
son to escape by spurring his horse, and the boy rode on to 
a place of safety, but turned and rode back deliberately to die 
with his father. 2 Potgieter drove back the Zulus after the fall 
of Uys, but he did not venture to hold his ground, and with- 
drew across the Drakensberg. Only a determined rally and 
crushing blow could free Natal from the hanging menace of the 
impis that Chaka had trained for the hand of Dingaan. 

In December, 1838, a force of six hundred mounted Boers 

was mustered to strike this blow under the command of Andries 

Pretorius. It seemed an absurdly weak force for such an attack, 

but the count in numbers did not measure its strength. Every 

1 "Annals of Natal," pp. 214-218. 2 Ibid. p. 374. 



THE PIONEER ADVANCE 107 

man was a master marksman with the heavy rifle that had so 
often broken the bound of the lion and stopped the charging 
rhinoceros when to miss was death. In every one's heart was 
a flame of hate for the ruthless Zulu. " Remember Retief " 
was a mutter that ran from man to man as the troop rode on. 
They longed for revenge as thirsty men crave water. They 
advanced, too, with the spirit of the Israelites of old and of 
Cromwell's Ironsides. They marched only between matins 
and evensong. They prayed in their saddles and lifted their 
voices in psalms. Surely the God of their covenant had the 
power to confound any might of the heathen and deliver their 
enemy into their hands. 

When they drew near to the Zulus, Pretorius halted, and 
with all his men offered a vow to the God of their fathers, 
should He grant them the victory, "to raise a house in memory 
of His great name wherever it should please Him, and note the 
day in a book to make it known to latest posterity." 1 

With this simple confidence in Divine protection there was 
the shrewdest practical judgment in selecting the best possible 
post to offset their comparative weakness in numbers and in- 
trench their little force. Their laager was pitched at the junc- 
tion of a broad river reach, called a sea-cow hole, with a deep, 
dry water-course, covering both flanks. Here, on Sunday, the 
1 6th of December, 1838, at five o'clock in the morning, they 
were attacked by a force of many thousand Zulus and fought 
for more than five hours. Impi after impi, reckless of life, 
charged up to the rifle front belching smoke, flame, and bullets, 
only to reel back before the deadly hail. When even this rag- 
ing horde wavered, Pretorius with one hundred and fifty picked 
horsemen circled about and struck their rear with a charge so fiery 
that the Zulus were utterly routed. The Boers drove the blacks 
to the river, shooting and trampling them under the feet of their 
horses. " The Kafirs lay on the ground," said one horseman, 
" like pumpkins in a rich soil that has borne a large crop." The 
sea-cow hole was packed so full that " the water looked like a 
1 " Annals of Natal," pp. 246-249, 448. 



io8 THE DIAMOND MINES OF SOUTH AFRICA 



pool of blood," and the stream thenceforward was known as 
Blood River. 1 Three thousand six hundred Zulus were left 
dead on the field, and this decisive victory was gained without 
the loss of a single life to the Boers. A few were slightly 
wounded, but they thought nothing of their hurts in the com- 
mon thanksgiving. 

This signal triumph and salvation were humbly taken as the 
answer of God to their prayers, and the vow before the battle 

was faithfully ful- 
filled, as the old 
Dutch Reformed 
Church of Pieter 
Maritzburg, the 
mother church 
of Southeast Af- 
rica, bears wit- 
ness. The flying 
Zulus were pur- 
sued and the 
kraal of Dingaan 
captured, Febru- 
ary jd, 1839, 
where the bodies 
of Retief and 
his companions 
were found and 
mournfully buried in one grave. The Boers called the place 
Weenan, the weeping, and so it is known to this day. 

Dingaan fled north and hid himself in a concealed kraal 
which he built. A Boer writer tells a story of his capture and 
death with grim delight. Many of the tribes which had been 
pressed in with the Zulus made peace with the Boers. One of 
the Swazi chiefs, Sapusa, who had bowed to the tyranny of 
Dingaan, found his late master's hiding-place. " On the first 
day old Sapusa pricked his captive with sharp assagais, not 

1 "Annals of Natal," pp. 246-249, 448. 




A Zulu Laborer in War Attire. 



THE PIONEER ADVANCE 109 

more than skin deep, from the sole of his foot to the top of his 
head. On the second day he caused him to be bitten by dogs. 
On the third day Sapusa said to Dingaan, ' Are you still the rain- 
maker, greatest of men ? The sun is rising, you shall not see 
it set.' Then he took assagais and bored Dingaan's eyes out, 
and when the sun set, Dingaan died, for he had had no food or 
water for three days. Such was the end of Dingaan." l 

So the Boers finally stayed the sweep of the Zulu scourge 
which had laid waste a great stretch of land north of the Cape 
settlements. Upon the defeat and flight of Umsilikazi, the vic- 
torious commandant, Hendrik Potgieter, proclaimed that all the 
territory overrun by this chief was forfeited to the pioneer 
Boers. This claim covered the greater part of the late South 
African Republic, and half, at least, of what is now the Orange 
River Colony. In this assertion there was no recognition of 
any sovereignty of Great Britain or attachment to the Cape 
Colony. It was the view of the Boers that the land which they 
took was theirs by right of capture and forfeit, and that they 
were independent adventurers with no ties of allegiance. A 
simple form of republican government was established for the 
Boers, north of the Orange River, by a general assembly of the 
pioneers at Winburg in June, 1837, and a few years later, on 
the land won from Dingaan, on the other side of the Drakens- 
berg, the republic of Natalia was declared to extend from the 
Umzimbulu to the Tugela. Outside of these crudely organized 
political associations there were from sixteen to twenty pioneer 
companies, headed by field cornets, which were practically as 
independent as the native tribes north of the Drakensberg. 
Neither of the republican creations was recognized by Great 
Britain, and, in 1842, Port Natal and the seaboard of the 
republic were captured, though Andries Pretorius repulsed the 
first British attack at Congella with heavy loss. In the follow- 
ing year Natal was formally declared to be a British Colony, and 
several thousand British immigrants were brought in to take the 

1 Of the basic fact of the assassination of Dingaan by a Swazi there is no 
question. 



no THE DIAMOND MINES OF SOUTH AFRICA 

place of the retiring Boers who recrossed the Drakensberg. In 
1848, by proclamation of Sir Harry Smith, her Majesty's High 
Commissioner and Governor of Cape Colony, all the territory 
between the Vaal and Orange rivers and the Quathlamba divi- 
sion of the Drakensberg was formally declared to be part of the 
British dominions under the name of the Orange River Sover- 
eignty. The Boers had been spreading out towards the Vaal in 
many trekking parties north of the Drakensberg, and the Brit- 
ish supremacy was not recognized until it was forcibly asserted 
by arms in the battle of Boomplatz, July 22, 1848. Then part 
of the Boers sullenly submitted, but many, headed by Andries 
Pretorius, preferred to pass beyond the farthest assertion of 
English dominion by crossing the Vaal and entering the wilder- 
ness stretching to the Limpopo. 

There was then not even a glimmer of anticipation that the 
great stretch of veld and karoo between the Orange and the Vaal 
contained by far the richest diamond fields in the world. The 
controlling ministry in Great Britain at the time did not even 
consider it worth the cost of keeping and defending, and on 
October 21, 1851, Earl Grey wrote to Sir Harry Smith that 
" its ultimate abandonment should be a settled point in imperial 
policy." The territory beyond the Vaal was rated still more 
cheaply, and on January 17, 1852, the local independence of 
the inhabitants of the Transvaal was formally recognized by the 
Sand River Convention, signed by two assistant commissioners 
for Sir Harry Smith, and by appointed delegates for the Trans- 
vaal pioneers. The state organization of these settlers was first 
christened Hollandsche Afrikaansche Republiek, but this name 
was changed to Zud Afrikaansche Republiek in September, 
1853. In the preceding month of July, Andries Pretorius, the 
pioneer leader who broke the Zulu power, died, but his great 
service was honorably recognized in the choice of his eldest son, 
Marthinus Wessel Pretorius, as the first president of the new 
Republic, and in the establishment of its capital of Pretoria. 

On March 31, 1852, Lieutenant General George Cathcart suc- 
ceeded Sir Harry Smith as High Commissioner and Governor 



THE PIONEER ADVANCE m 

of Cape Colony. The Transvaal had been already disposed of 
by the Sand River Convention, but, immediately after his arrival, 
May 13, 1852, General Cathcart issued a formal proclamation 
confirming this convention. It appeared, too, that it might be 
desirable to shift the charge of maintenance and local defence of 
the Orange River Sovereignty to the shoulders of the pioneer 
settlers. This conviction was confirmed by the outbreak of a 
war with the Basutos, the most powerful native tribe in this terri- 
tory, under a cunning chief, Moshesh. In November, 1852, Gen- 
eral Cathcart led a little army of two thousand infantry and five 
hundred cavalry to the Caledon River, but in the following month 
his expedition was beset by an overwhelming force of Basutos 
at Berea Mountain, and the battle was in effect a repulse to 
the British. After leaving a garrison at Bloemfontein, General 
Cathcart withdrew under cover of a fragile proclamation of 
peace, but his report and the accompanying news were so dis- 
couraging that the Duke of Newcastle wrote to him that " her 
Majesty's Government had decided to withdraw from the 
Orange River Sovereignty." In pursuance of this conclusion 
a convention was signed February 23, 1854,31 Bloemfontein, by 
Sir George Russell Clerk, special commissioner representing 
Great Britain, and by the delegates from districts in the sov- 
ereignty. By this convention the independence of the settlers 
in the sovereignty was guaranteed, and the administration was 
handed over to a provisional council, which took charge until 
the first sitting of the Volksraad, March 28, 1854, and the 
declaration of a republic in the following month under the name 
of the Orange Free State. This independent state covered the 
greater part of the territory comprised within the bounds of the 
Orange River Sovereignty, excepting the large division between 
the Caledon River and the Quathlamba Mountains, reserved 
to the Basutos, and smaller reservations on the Vaal held by 
the Griquas. 

Within the limits of the whole district between the Orange 
and the Vaal rivers there were then not more than fifteen thou- 
sand whites scattered over a territory of many thousand square 



ii2 THE DIAMOND MINES OF SOUTH AFRICA 

miles. Except in the Caledon River districts little of this great 
expanse was capable of supporting any clustered population or 
even available for agriculture. The soil throughout was shallow, 
and in the southern and western sections the rainfall was ordi- 
narily light. There were a number of widespreading karoos, and 
in the dry months the greater part of the veld was little better 
than the desert. The so-called farms were chiefly cattle and sheep 
pastures, where the yield of grass and herbage was so varying 
that several thousand acres were needed for any fair assurance of 




safety for a small herd. The total number of farms secured by 
grant was only twelve hundred and sixty-five, but they extended 
over eleven million acres. Of the farm owners only one hundred 
and thirty-nine were Englishmen, and a number of these were non- 
residents. 1 In the abstract there was seemingly little attraction or 
value to excite any flow of immigration or to make the province 
a prize worth the cost of defending. 

Not only the prospects of the Orange Free State and of its 
neighbor on the other side of the Vaal seemed dull and incon- 
siderable to most observers, but the condition of Natal and 

i " South Africa," Theal. 



THE PIONEER ADVANCE 113 

of Cape Colony itself was little more promising. In Great 
Britain the whole dependency was so lightly esteemed that it 
was determined in 1849 to utilize it as a dumping ground for 
convicts, after Australia had resentfully thrown off this burden. 
The convict ship Neptune was actually sent out, but the indigna- 
tion of the colonists was so demonstrative that no convicts were 
landed, and the ship with its load was held for five months in 
Simon's Bay, the present Naval Station, a little south of Cape 
Town, until the recalling order was received, February 13, 1850. 
The colony had not sunk so low as to submit to this mark of 
contempt, but it was undoubtedly drooping in hopes and enter- 
prise, and the progress of its industrial development was pain- 
fully slow. There had been a pronounced diversion from 
agriculture to cattle and sheep raising for reasons before noted, 
and wool had become the chief and almost the only export of 
consequence. Still the peculiar condition and vagaries of the 
South African climate and seasons were hard to provide for or 
overcome, and there were prevalent diseases that attacked horses, 
cattle, and sheep, and greatly checked the rise of the pastoral 
industry. Communication from one part of the colony to 
another was very slowly improved. The roads were few and 
bad, and in 1867 the only stretch of railway in all South Africa 
was a bare forty miles from Cape Town to Wellington. The 
total annual export of the Colony was a trifle over ^2,000,000 
in value, and there was no diversification of industries and no 
manufactures of any considerable extent. 1 This was the situation 
when the gloom was suddenly dispelled and the whole face of 
South Africa changed by the discovery of the Diamond Fields. 

1 " South Africa," Theal. 




CHAPTER IV 

THE DISCOVERY 

EARLY two hundred years had passed since 
the memorable expedition of van der Stel 
made known to geographers the Groote River, 
which, a hundred years later, was christened 
the Orange. Before Great Britain took the 
Cape, the daring van Reenen had penetrated 
to Modder Fontein, unconsciously skirting the rim of a marvel- 
lous diamond field. Since the beginning of the century scores 
of roving hunters had chased their game over a network of 
devious tracks, traversing every nook of the land between the 
Orange and the Vaal, and often camping for days upon their 
banks. Then the trekking pioneer graziers and farmers plodded 
on after the hunters, sprinkling their huts and kraals over the 
face of the Orange Free State, but naturally squatting first on 
the arable lands and grazing ground nearest the water-courses. 
So, in the course of years, in the passage of the Great Trek, 
thousands of men, women, and children had passed across the 
Orange and Vaal, and up and down their winding valleys, and 
hundreds, at least, had trodden the river shore sands of the 
region in which the most precious of gems were lying. 

On the Orange River, some thirty miles above its junction 
with the Vaal, there was the hamlet of Hopetown, one of the 
most thriving of the little settlements, and a number of farms 
dotted the angle between the rivers. Along the line of the Vaal, 
for some distance above its entry into the Orange, there were 
some ill-defined reservations occupied by a few weak native 
tribes, Koranas and Griquas, for whose instruction there 

114 



THE DISCOVERY 115 

were mission stations at Pniel and Hebron. 1 For centuries 
unnumbered the aboriginal tribes had been ignorantly trampling 
under foot gems of countless price, and for years Dutch and 
English hunters, pioneers, farmers, shepherds, and missionaries 
trekked as heedlessly over the African diamond beds. 

After the revelation of this fact, there arose, it is true, an 
imposing tale of an old mission map of the Orange River region, 
drawn as far back as the middle of the eighteenth century, across 
whose worn and soiled face was scrawled: "Here be diamonds."' 2 
Even if this report were true, there was no evidence determining 
the date of the scrawl, which might more credibly be a crude 
new record than a vague old one. In any event, it does not 
appear that there was even a floating rumor of the probable 
existence of a South African diamond field at the time of the 
actual discovery of the first identified gem. 

There is nothing surprising in this oversight. When a 
spectator beholds a great semicircle of artfully cut gems spar- 
kling on the heads, necks, and hands of fair women massed in 
superb array, and resplendent in the brilliant lights of an opera 
house, or when one views the moving throng glittering with 
jewels in grand court assemblies, it is hard for him to realize 
how inconspicuous a tiny isolated crystal may be in the richest 
of earth beds. No spot in a diamond field has the faintest 
resemblance to a jeweller's show tray. Here is no display of 
gems blazing like a Mogul's throne, or a Queen's tiara, or the 
studded cloak of a Russian noble. Only in the marvellous 
valley of Sindbad are diamonds strewn on the ground in such 
profusion that they are likely to stick in the toes of a barefooted 
traveller, and can be gathered by flinging carcasses of sheep from 
surrounding precipices to tempt eagles to serve as diamond 
winners. 

It needs no strain of faith to credit the old Persian tale of 
the discontented AH Hafed, roaming far and wide from his 

1 " South Africa," George McCall Theal, London, 1888, 1891, 1893. 

2 " South African Diamond Fields and Journey to Mines," William Jacob 
Morton, New York, 1877. 



n6 THE DIAMOND MINES OF SOUTH AFRICA 

charming home on the banks of the Indus in search of dia- 
monds, and, finally, beggared and starving, casting himself into 
the river which flowed by his house, while the diamonds of Gol- 
conda were lying in his own garden sands. It is probable that 
the diamonds of India were trodden under foot for thousands of 
years before the first precious stone of the Deccan was stuck in 
an idol's eye or a rajah's turban. It is known that the Brazilian 
diamond fields were washed for many years by gold placer 
diggers without any revelation of diamonds to the world, 
although these precious stones were often picked up and so 
familiarly handled that they were used by the black slaves in the 
fields as counters in card games. 

If this be true of the most famous and prolific of all dia- 
mond fields before the opening of the South African placers and 
mines, any delay in the revelation of the field in the heart of 
South Africa may be easily understood. For it was not only 
necessary to have eyes bright and keen enough to mark one of 
the few tiny precious crystals which were lying on the face of 
vast stretches of pebbles, boulders, and sand, but the observer 
must prize such a crystal enough to stoop to pick it up if it lay 
plainly before his eyes. To the naked native a rough diamond 
had no more attraction than any other pretty pebble. There 
were millions of other white crystals and many colored pebbles 
on the river shores which were equally precious or worthless in 
his eyes. The roving hunters were looking sharply for game 
bounding over the veld, and only glanced at a pebble-strewn 
bank to mark the possible track of their prey. The stolid Boer 
pioneers would hardly bend their backs to pick up the prettiest 
stone that ever lay on the bank of an African river, even if it 
were as big as the great yellow diamond so jealously guarded by 
the Portuguese crown. 1 

It might be thought that some visitor to the fields would be 
more expert in judging its character than natives, hunters, and 
farmers ; but there were few trained mineralogists in South 

1 "The Gold Regions of Southeastern Africa," Thomas Baines, F.R.G.S., 
London, 1877. 



THE DISCOVERY 117 

Africa, and it is doubtful if there was one who had ever examined 
a diamond field personally or compared one field with another. 
Even with this special experience an expert student of general 
mineral formations might survey this particular field closely with- 
out suspecting the existence of diamonds. This was demon- 
strated in the visit of the colonial geologist Wyley to the 
Orange Free State in 1856, when he investigated the alleged 
discovery of gold in thin veins of quartz lining the joints and 
crevices of the trappean rocks at Smithfield. In the course of 
his exploration he went to Fauresmith, where diamonds were 
afterward picked from the town commonage, and stood on the 
verge of the farm Jagersfontein, later the seat of a prolific dia- 
mond mine, yet it does not appear that he had even a surmise 
of the existence of diamonds in the field of his investiga- 
tion. 1 It is but fair to him to observe, however, that the sec- 
tion which he visited had no such close resemblance to any 
known typical field as that which led Humboldt and Rose to 
the revelation of the diamonds of the Ural from the similarity 
of the ground formations to those of the Brazilian diamond 
districts. 

As a matter of fact nobody who entered the Vaal river region 
conceived it to be a possible diamond field or thought of search- 
ing for any precious stones. Probably, too, there was not a 
person in the Orange Free State, and few in the Cape Colony, 
who was able to distinguish a rough diamond if he found one by 
chance, or would be likely to prize such a crystal. For the dis- 
covery of diamonds under such conditions it was practically 
necessary that a number of prospectors should enter it who 
would search the gravel beds often and eagerly for the prettiest 
pebbles. Were any such collectors at work in the field ? 

One of the trekking Boers, Daniel Jacobs, had made his 
home on the banks of the Orange River near the little settle- 
ment of Hopetown. He was one of the sprinkling of little 
farmers who was stolidly content with a bare and precarious liv- 

1 "Among the Diamonds," by the late John Noble, Clerk of the House of 
Assembly, Cape Town. 



u8 THE DIAMOND MINES OF SOUTH AFRICA 

ing on the uncertain pasture lands of the veld. Here his chil- 
dren grew up about him with little more care than the goats that 
browsed on the kopjes. 

A poor farmer's home was a squalid hovel. It was roughly 
partitioned to form a bedroom and kitchen, lighted by two small 
windows smudged with grime. Dirty calico tacked on the 
rafters made its ceiling. Its bare earthen floor was smeared 
weekly with a polishing paste of cowdung and water. Father, 
mother, and children slept together on a rude frame overlaced 
with rawhide strips. The only other furniture in this stifling 
bedroom was a chest of drawers and a small cracked mirror. 
There was no washbowl or water pitcher, but in the morning 
one after another of the family wiped their faces and swabbed 
their hands on the same moistened cloth. Then they drew up 
chairs with rawhide seats to a rough wooden table and ate corn 
meal porridge, and sometimes a hunk of tough mutton boiled 
with rice, and soaked their coarse unbolted wheat flour bread in 
a gritty, black coffee syrup. 1 

When the sheep and goats were turned out of the kraal to 
graze on the patches of grass and the stunted thorns of the veld, 
the children ran away after them and roamed over the pasture 
land all day long like the flocks. There was no daily round of 
work for them. The black servants were the shepherds of the 
flocks, and did the slovenly housework, under the indolent eye 
of the Boer and his vrouw, for the poorest farmer would not 
work with his own hands except at a pinch. His boys and girls 
had never seen a doll or a toy of any kind, but the instinct of 
childhood will find playthings on the face of the most barren 
karoo, and the Jacobs children were luckily close to the edge of 
a river which was strewn with uncommonly beautiful pebbles, 
mixed with coarser gravel. 

Here were garnets with their rich carmine flush, the fainter 
rose of the carnelian, the bronze of jasper, the thick cream of 
chalcedony, heaps of agates of motley hues, and many shining 

1 " Life with the Boers in the Orange Free State," by a resident English physi- 
cian's wife, New York, 1899. 



THE DISCOVERY 119 

rock crystals. 1 From this party-colored bed the children picked 
whatever caught their eye and fancy, and filled their pockets with 
their chosen pebbles. So a poor farmer's child found playthings 
scattered on a river bank which a little prince might covet, and 
the boy might have skimmed the face of the river with one litcle 
white stone that was worth more than his father's farm. Fortu- 
nately for the future of South Africa, he did not play ducks and 
drakes with this particular stone, which he found one day in the 
early spring of 1867, but carried it home in his pocket and 
dropped it with a handful of other pebbles on the farmhouse 
floor. 2 

A heap of these party-colored stones was so common a sight 
in the yard or on the floor of a farmhouse on the banks of the 
Orange and Vaal, that none of the plodding Boers gave it a 
second glance. But when the children tossed the stones about, 
the little white pebble was so sparkling in the sunlight that it 
caught the eye of the farmer's wife. She did not care enough 
for it to pick it up, but spoke of it as a curious stone to a neigh- 
bor, Schalk van Niekerk. Van Niekerk asked to see it, but it 
was not in the heap. One of the children had rolled it away in 
the yard. After some little search it was found in the dust, for 
nobody on the farm would stoop for such a trifle. 

When van Niekerk wiped off" the dust, the little stone glit- 
tered so prettily that he offered to buy it. The good vrouw 
laughed at the idea of selling a pebble. " You can keep the 
stone, if you want it," she said. So van Niekerk put it in his 
pocket and carried it home. He had only a vague notion that 
it might have some value, and put it in the hands of a travelling 
trader, John O'Reilly, who undertook to find out what kind of 
a stone the little crystal was, and whether it could be sold. He 

1 "The Diamond Diggings of South Africa," Charles Alfred Payton, Lon- 
don, 1872. "South Africa Diamond Fields," Morton, New York, 1877. 
"Diamonds and Gold of South Africa," Henry Mitchell of Kimberley, London, 
1889. 

2 "Among the Diamonds," 1870-1871. "South Africa," Theal, London, 
1888-1893. 



120 THE DIAMOND MINES OF SOUTH AFRICA 



showed the stone to several Jews in Hopetown and in Coles- 
berg, a settlement farther up the Orange River Valley. No 
one of these would give a penny for it. " It is a pretty stone 
enough," they said, "probably a topaz, but nobody would pay 
anything for it." 

Perhaps O'Reilly would have thrown the pebble away, if it 
had not come under the eye of the acting Civil Commissioner 
at Colesberg, Mr. Lorenzo Boyes. Mr. Boyes found on trial 
that the stone would scratch glass. 

" I believe it to be a diamond," he observed gravely. 1 
O'Reilly was greatly cheered up. "You are the only man 

I have seen," he said, " who 
says it is worth anything. 
Whatever it is worth you 
shall have a share in it." 

" Nonsense," broke in 
Dr. Kirsh, a private apothe- 
cary of the town, who was 
present, " I'll bet Boyes a 
new hat it is only a topaz." 
" I'll take the bet," re- 
plied Mr. Boyes, and at his 
suggestion the stone was 
sent for determination to the 
John o'Reiiiy. foremost mineralogist of the 

colony, Dr. W. Guybon Atherstone, residing at Grahamstown. 
It was so lightly valued that it was put in an unsealed envelope 
and carried to Grahamstown in the regular post-cart. 

When the post-boy handed the letter to Dr. Atherstone, 
the little river stone fell out and rolled away. The doctor 
picked it up and read the letter of transmission. 2 Then he 
examined the pebble expertly and wrote to Mr. Boyes : " I 
congratulate you on the stone you have sent to me. It is a 

1 Lorenzo Boyes (statement furnished to author), 1899. 

2 W. Guybon Atherstone ; Lorenzo Boyes, 1899. " Among the Diamonds," 
1870-1871. 




THE DISCOVERY 



121 




veritable diamond, weighs twenty-one and a quarter carats, and 
is worth ^500. It has spoiled all the jewellers' files in Grahams- 
town, and where that came from there 
must be lots more. Can I send it to 
Mr. Southey, Colonial Secretary ? " 

This report was a revelation which 
transformed the despised Karooland as 
the grimy Cinderella was transfigured 
by the wand of her fairy godmother. 
The determination was so positive and 
the expertness of the examiner so well 
conceded that Sir Philip Wodehouse, 
the Governor at the Cape, bought the Mr - Lorenzo Boyes - 

rough diamond at once, at the value fixed by Dr. Atherstone and 
confirmed by the judgment of M. Henriette, the French consul 
in Cape Town. 1 The stone was sent immediately to the Paris 
Exhibition, where it was viewed with much interest, but its dis- 
covery, at first, did not cause any great sensation. The occa- 
sional finding of a diamond in a bed of pebbles had been 
reported before from various parts of the globe, and there was 

no assurance in this discovery of any 
considerable diamond deposits. 

Meanwhile Mr. Boyes hastened 
to Hopetown and to van Niekerlc's 
farm, to search along the river shore 
where the first diamond was found. 
He prodded the phlegmatic farmers 
and their black servants, raked over 
many bushels of pebbles for two 
weeks, but no second diamond repaid 
his labor. Still the news of the find- 
Dr. w. Guybon Atherstone. ing of the first stone made the farmers 

near the river look more sharply at every heap of pebbles in the 
hope of finding one of the precious " blink klippe " (bright stones), 

1 "South Africa," Theal. Lorenzo Boyes, 1899. "Diamonds and Gold 
of South Africa," Theodore Reunert, 1893. 




122 THE DIAMOND MINES OF SOUTH AFRICA 

as the Boers named the diamond, and many bits of shining rock 
crystal were carefully pocketed, in the persuasion that the glit- 
tering stones were diamonds. But it was ten months from the 
time of the discovery at Hopetown before a second diamond 
was found, and this was in a spot more than thirty miles away, 
on the river bank below the junction of the Vaal and Orange 
rivers. Mr. Boyes again hastened to the place from which the 
diamond had been taken, but he failed again to find companion 
stones, though he reached the conclusion that the diamond had 
been washed down stream by the overflowing of the Vaal. 1 

From the Orange River the search passed up the Vaal, where 
the beds of pebbles were still more common and beautiful. The 
eyes of the native blacks were much quicker and keener in 
such a quest than those of the stolid Boer, who scarcely troubled 
himself to stoop for the faint chance of a diamond. But no 
steady or systematic search was undertaken by anybody, and it 
was not until the next year, 1868, that a few more diamonds were 
picked up on the banks of the Vaal by some sharp-sighted 
Koranas. 2 The advance of discovery was so slow and disap- 
pointing that there seemed only a faint prospect of the realization 
of the cheering prediction of Dr. Atherstone, which was scouted 
by critics who were wholly incompetent to pass upon it. Even 
the possibility of the existence of diamond deposits near the 
junction of the Orange and Vaal was flatly denied by a preten- 
tious examiner who came from England to report on the Hope- 
town field. It was gravely asserted that any diamonds in that 
field must have been carried in the gizzards of ostriches from 
some far-distant region, and any promotion of search in the 
field was a bubble scheme. 

To this absurd and taunting report Dr. Atherstone replied 
with marked force and dignity, presenting the facts indicating 
the existence of diamond-bearing deposits, and adding : " Suf- 
ficient has been already discovered to justify a thorough and 
extensive geological research into this most interesting country, 
and I think for the interest of science and the benefit of the 
1 Lorenzo Boyes, 1899. "South Africa," Theal, London, 1888-1893. 



THE DISCOVERY 123 

Colony a scientific examination of the country will be under- 
taken. So far from the geological character of the country mak- 
ing it impossible, I maintain that it renders it probable that very 
extensive and rich diamond deposits will be discovered on proper 
investigation. This I trust the Home Government will author- 
ize, as our Colonial exchequer is too poor to admit of it." 1 

There was no official response to this well-warranted sug- 
gestion, for it had hardly been penned when the announcement 
of a remarkable discovery aroused such an excitement and such 
a rush to the field that no government exploration was needed. 
In March, 1869, a superb white diamond, weighing 83.5 carats, 
was picked up by a Griqua shepherd boy on the farm Zendfon- 
tein, near the Orange River. 2 Schalk van Niekerk bought this 
stone for a monstrous price in the eyes of the poor shepherd, 
500 sheep, 10 oxen, and a horse, but the lucky purchaser sold 
it easily for ^11,200 to Lilienfeld Brothers of Hopetown, and 
it was subsequently purchased by Earl Dudley for _^25,ooo. 3 
This extraordinary gem, which soon became famous as " the 
Star of South Africa," drew all eyes to a field which could yield 
such products, and the existence and position of diamond beds 
was soon further assured and defined by the finding of many 
smaller stones in the alluvial gravel on the banks of the Vaal. 

Alluvial deposits form the surface ground on both sides of 
this river, stretching inland for several miles. In some places 
the turns of the stream are frequent and abrupt, and there are 
many dry water-courses which were probably old river channels. 
The flooding and winding of the river partly accounts for the 
wide spreading of the deposits, but there has been a great abrasion 
of the surface of the land, for the water-worn gravel sometimes 
covers even the tops of the ridges and kopjes along the course 
of the river. 

This gravel was a medley of worn and rolled chips of basalt, 
sandstone, quartz, and trap, intermingled with agates, garnets, 

1 W. Guybon Atherstone, 1868. 2 " Among the Diamonds," 1870-1871. 
3 Ibid. (Accounts of this discovery differ somewhat.) Vide Theal's "South 
Africa," Reunert's " Diamonds and Gold," etc. 



124 THE DIAMOND MINES OF SOUTH AFRICA 

peridot, jasper, and other richly colored pebbles, lying in and on 
a bedding of sand and clay. Below this alluvial soil was in some 
places a calcareous tufa, but usually a bed rock of melaphyre or 
a clayey shale varying in color. Scattered thickly through the 
gravel and the clay along the banks were heavy boulders of 
basalt and trap which were greatly vexing in after days to the 
diamond diggers. 1 

For a stretch of a hundred miles above the Mission Station 
at Pniel the river flows through a series of rocky ridges, rolling 
back from either bank to a tract of grassy, undulating plains. 
Fancy can scarcely picture rock heaps more contorted and mis- 
shapen. Only prodigious subterranean forces could have so 
rent the earth's crust and protruded jagged dykes of metamor- 
phic, conglomerate, and amygdaloid rocks, irregularly traversed 
by veins of quartz, and heavily sprinkled with big bare boulders 
of basalt and trap. Here the old lacustrine sedimentary forma- 
tion of the South African high veld north of the Zwarte Bergen 
and Witte Bergen ranges has plainly been riven by volcanic 
upheaval. The shale and sandstone of the upper and lower 
Karoo beds have been washed away down to an igneous rock 
lying between the shale and the sandstone. It was along this 
stretch of the river that the first considerable deposit of diamonds 
in South Africa was uncovered. 2 

For more than a year since the discovery of the first diamond 
there had been some desultory scratching of the gravel along the 
Vaal by farmers and natives in looking for " blink klippe," and 
a few little rough diamonds had been found by the Hottentots, 
as before noted ; but the first systematic digging and sifting of 
the ground was begun by a party of prospectors from Natal at 
the Mission Station of Hebron. This was the forerunner of the 

'"Diamonds and Gold of South Africa," Reunert, Cape Town, 1893. 
"The Diamond Diggings of South Africa," Payton, 1872. "Among the 
Diamonds," 1870-1871. 

2 " Diamonds and Gold of South Africa," Reunert, 1893. " Among the 
Diamonds," 1870-1871. "South Africa," Theal, 1888-1893. "On 
Diamonds," Sir William Crookes, London, 1897. 



THE DISCOVERY 125 

second Great Trek to the Vaal from the Cape, a myriad of 
adventurers that spread down the stream like a locust swarm, 
amazing the natives, worrying the missionaries, and agitating the 
pioneer republics on the north and the east. 1 

The first organized party of prospectors at Hebron on the 
Vaal was formed at Maritzburg in Natal, at the instance of 
Major Francis, an officer in the English army service, then 
stationed at that town. Captain Rolleston was the recog- 
nized leader, and after a long plodding march over the Drakens- 
berg and across the veld, the little company reached the valley 
of the Vaal in November, 1869. Up to the time of its arrival 
there had been no systematic washing of the gravel edging the 
river. Two experienced gold diggers from Australia, Glenie 
and King, and a trader, Parker, had been attracted to the 
field like the Natalians by the reported discoveries, and were 
prospecting on the line of the river when Captain Rolleston's 
party reached Hebron. 2 Their prospecting was merely looking 
over the surface gravel for a possible gem, but the wandering 
Koranas were more sharp-sighted and lucky in picking up the 
elusive little crystals that occasionally dotted the great stretches 
of alluvial soil. 

It was determined by Captain Rolleston to explore the 
ground as thoroughly as practicable from the river's edge for a 
number of yards up the bank, and the washing began on a tract 
near the Mission Station. The Australian prospectors joined 
the party, and their experience in placer mining was of service 
in conducting the search for diamonds. The workers shovelled 
the gravel into cradles, like those used commonly in Australian 
and American placer washing, picked out the coarser stones by 
hand, washed away the sand and lighter pebbles, and saved the 
heavier mineral deposit, hoping to find some grains of gold as 
well as diamonds above the screens of their cradles. But the 
returns for their hard labor for many days were greatly disap- 
pointing. They washed out many crystals and brilliant pebbles, 

1 " South Africa," Theal, 1888-1893. Among the Diamonds," 1870- 
1871. 2 "Among the Diamonds," 1870-1871. 



126 TIIK DIAMOND MINKS OK SOUTH AKRICA 

hut never a diamond nor an atom of gold dust. Then they 
pushed down the river more than twenty miles to another camp 
at Klip-drift, opposite the Mission Station at 1'niel. Here too 
they washed the ground for days without finding even the tiniest 
gem, and were almost on the point ot abandoning their dishearten- 
ing drudgery, when finally, on the seventh of January, 1870, the 
first reward of systematic work in the field came in the appear- 
ance of a small diamond in one of the cradles. 1 

This little fillip of encouragement determined their continu- 
ance of the work, and a party from British KarFraria joined 
them in washing the gravel in places that seemed most promis- 
ing along the line of the river. It was agreed that the first 
discovery of rich diamond-bearing ground should be shared 
alike by both parties, but there was nothing to share for 
some weeks. Then some native Koranas were induced to 
point out to the Natalians a gravel-coated hummock or kopje 
near the Klip-drift camp, where they had picked up some small 
diamonds. When the prospectors began the washing of the 
gravel on this kopje, it was soon apparent that a diamond bed 
of extraordinary richness had been reached at last. Good faith 
was kept with the company from Kingwilliamstown, and the 
combined parties worked to the top of their strength in shovel- 
ling and washing the rich bed. The lucky men kept their 
mouths closed, as a rule, and did not intend to make known their 
good fortune ; but such a discovery could not long be concealed 
from visiting traders and roaming prospectors, and before three 
months had passed some prying eye saw half a tumblerful of the 
white sparkling crystals in their camp, and the news spread fast 
that the miners had washed out from two hundred to three hun- 
dred stones, ranging in size from the smallest gems to diamonds 
of thirty carats or more. 3 

Then a motley throng of fortune-hunters began to pour into 

the valley of the Vaal. The first comers were those living 

nearest to the new diamond field, farmers and tradesmen from 

the cattle ranges and little towns of the Orange Free State. 

1 " Among the Diamonds," 1870-1871. * ItiJ. 



THK D1SCOVKRY 127 

Some of these were stolid Boers, drawn to the fields as a novel 
and curious spectacle, but disdaining the drudgery of shovelling 
and washing from morning till night for the chance of a tiny 
bright stone. They stared for a while at the laboring diamond 
seekers, and then turned their backs on the field contemptuously, 
and rode home sneering at the mania which was dragging its 
victims for hundreds of miles, over sun-cracked and dusty 
karoos, to hunt for white pebbles in a river bed. Still there 
were many poor farmers who caught the infectious diamond 
fever at sight of the open field and a few sparkling stones, and 
they camped at Klip-drift or went on farther up or down the 
river, to join, as well as they knew how, in the search for 
diamonds. 

Following this influx from the Free State came swarming 
in men of every class and condition from the southern English 
Colony, and from the ships lying in the coast ports. The 
larger number were of English descent, but many were Dutch, 
and hardly a nation in Europe was unrepresented. Black 
grandsons of Guinea coast slaves and natives of every dusky 
shade streaked the show of white faces. Butchers, bakers, 
sailors, tailors, lawyers, blacksmiths, masons, doctors, carpenters, 
clerks, gamblers, sextons, laborers, loafers, men of every pur- 
suit and profession, jumbled together in queerer association than 
the comrades in the march to Finchley, fell into line in a 
straggling procession to the Diamond Fields. Army officers 
begged furloughs to join the motley troop, schoolboys ran away 
from school, and women even of good families could not be held 
back from joining their husbands and brothers in the long and 
wearisome journey to the banks of the Vaal. 1 

There was the oddest medley of dress and equipment: shirts 
of woollen, blue, brown, gray, and red, and of linen and 
cotton, white, colored, checked, and striped ; trim jackets, cord 
riding-breeches and laced leggings, and " hand me downs " from 
the cheapest ready-made clothing shops ; the yellow oilskins 
and rubber boots of the sailor; the coarse, brown corduroy and 

1 "Among the Diamonds," 1870-1871. 



THE DIAMOND MINES OF SOUTH AFRICA 

canvas suits, and long-legged, stiff, leather boots of the miner; 
the ragged, greasy hats, tattered trousers or loin cloths of the 
native tribesmen ; jaunty cloth caps, broad-brimmed felt, bat- 
tered straw, garish handkerchiefs twisted close to the roots of 
stiff black crowns, or tufts of bright feathers stuck in a wiry mat 
of curls ; such a higgledy-piggledy as could only be massed in 
a rush from African coast towns and native kraals to a field of 
unknown requirements, in a land whose climate swung daily be- 
tween a scorch and a chill, where men in the same hour were 
smothered in dust and drenched in a torrent. 

It is doubtful if a single one of this fever-stricken company 
had ever seen a diamond field or had the slightest experience 
in rough diamond winning, but no chilling doubt of them- 
selves or their luck restrained them from rushing to their 
fancied Golconda. Their ideal field was much nearer a mirror 
of the valley of Sindbad than the actual African river bank, and 
it was certain that many would be as bitterly disappointed by 
the rugged stretch of gravel at Klip-drift as the gay Portuguese 
cavaliers were at the sight of the Manica gold placers. 

Everything in the form of a carriage from a chaise to a buck- 
wagon was pressed into service, but the best available transport 
was the big trekking ox-wagon of the Boer pioneer. This was a 
heavily framed, low-hung wagon, about twenty feet long and five 
and a half feet broad. In this conveyance more than a dozen 
men often packed themselves and their camping outfit and food. 
An exceptionally well-equipped party carried bacon, potatoes, 
onions, tea, coffee, sugar, condensed milk, flour, biscuits, dried 
peas, rice, raisins, pickles, and Cape brandy. The total weight 
of load allowed, including the living freight, was limited to 
seven thousand pounds. 1 

East London, the nearest port, was something more than 
four hundred miles from the diamond field, and Cape Town 
nearly seven hundred. Natal, Port Alfred, and Port Elizabeth 
were almost equally distant, as the crow flies, approximately four 

1 "The Diamond Diggings of South Africa," Charles Alfred Payton, Lon- 
don, 1872. 



THE DISCOVERY 129 

hundred and fifty miles ; but the length of the journey to the 
Vaal could not be measured by any bare comparison of air-lined 
distances. The roads, at best, were rough trampled tracks, 
changing, after a rainfall, to beds of mire. Their tortuous 
courses rambled from settlement to settlement, or from one 
farmhouse to another over the veld, and were often wholly lost 
in the shifting sands of the karoo. It was a tedious and diffi- 
cult journey by land even from one seacoast town to another, 
and fifty miles from the coast the traveller was fortunate if his 
way was marked by even a cattle path. 1 

When the rain fell in torrents with the lurid flashes and 
nerve-shaking crash of South African thunder-storms, the dia- 
mond seekers huddled together under the stifling cover of their 
wagons, while fierce gusts shook and strained every strip of 
canvas and water drops spurted through every crevice. In fair 
weather some were glad to spread their blankets on the ground 
near the wagon, and stretch their limbs, cramped by their pack- 
ing like sardines in a box. On the plains they had no fuel for 
cooking except what they could gather of dry bullock's dung. 
Sometimes no headway could be made against the blinding dust- 
storms, that made even the tough African cattle turn tail to the 
blasts, and clogged the eyes and ears and every pore of exposed 
skin with irritating grit and powder. Sometimes the rain fell so 
fast that the river beds were filled in a few hours with muddy 
torrents, which blocked any passage by fording for days and 
even weeks at a time, and kept the impatient diamond seekers 
fuming in vain on their banks. Payton's party was forty-six 
days in its passage from Port Elizabeth to the Diamond Fields 
without meeting with any serious delays, and journeys lasting 
two months were not uncommon. 2 

Still, in spite of all obstacles, privations, and discomforts, the 
long journey to the fields was not wholly monotonous and un- 
pleasant. As there was no beaten way, the prospectors chose 

1 "South Africa," George McCall Theal, 1888-1893. 

2 " The Diamond Diggings of South Africa," Payton. " South Africa Dia- 
mond Fields and Journey to Mines," William Jacob Morton, New York, 1877. 



130 THE DIAMOND MINES OF SOUTH AFRICA 

their own path, riding by day and camping at night as their 
fancy led them. In ascending to the tableland of the interior 
from Natal, there were shifting and stirring visions of mountain 
peaks, terraces, gorges, and valleys. 

On the higher terraces there was not the luxuriance of the 
coast, the huge tree ferns with feathery fronds, the towering 
masses of palms, the drooping festoons of climbing vines, the 
exquisite flowers : spiked ansellias with their pale yellow blos- 
soms, barred and spotted with red, pure white, sweet-scented 
clusters of mystacidium, and orchids of marvellous variety and 
hue, but even the highest upland tree growth had beauties of 
its own. On the slopes of the Drakensberg the wild chestnut, 
the Natal mahogany, the white pear and iron wood grow sturdily, 
and the common yellow wood, stink wood, bogabog, and sneeze 
wood flourish in spite of their rude names. 1 

Amid this varied scenery they could linger and wind about 
as they pleased, and every turn of their path revealed new charms 
of line and color. As they descended the mountain flanks some 
marked how the lacustrine deposits of past ages had overspread 
the face of the land with their covering of sandstone and shale, 
even skirting the summits of the highest peaks at a height of 
more than six thousand feet, as was plainly shown on the Com- 
passberg. 2 On the plateau below they saw how the craggy hills, 
pointed spitz-kopjes, and columnar ridges of the trappean rocks 
projected above the sedimentary cover of the karoo. 

Throughout the Orange Free State, but especially in the 
neighborhood of the valleys of the Orange and Vaal, these vol- 
canic rock elevations are common, sometimes massed in irregular 
rows and often rising in the most jagged and fantastic shapes. 
" When we see them at the surface," wrote the geologist Wyley 
in 1856, "they look like walls running across the country, or 
more frequently form a narrow, stony ridge like a wall that 
has been thrown down. The rock of which they are composed, 
greenstone or basalt, is known by the local name of iron stone, 

1 "The Colony of Natal," J. Forsyth Ingram. 

2 "Among the Diamonds," 1870-1871. 



THE DISCOVERY 131 

from its great hardness and toughness, and from its great weight. 
The origin of these dykes is well known. They have been pro- 
duced by volcanic agency, which, acting from below upon hori- 
zontal beds of stratified rock, has cracked and fissured them at 
right angles to their planes of stratification, and these vertical 
cracks have been filled up with the melted rock or the lava from 
below. The perpendicular fissures through which it has found 
its way upwards are seldom seen, nor should we expect to see 
much of them, for it is precisely along the line of these that the 
rocks have been most broken up and shattered and the denuda- 
tion has been greatest." 

Even in the crossing of the karoos there were curious and 
awesome sights to attract and impress the mind of a traveller 
beholding for the first time these desert wastes so widely spread 
over the face of South Africa. They differ little in appearance 
except in size. The Great or Central Karoo, which lies beneath 
the foot-hills of the Zwarte Bergen range, has a sweep to the 
north of more than three hundred miles in a rolling plateau, 
ranging in elevation from two to three thousand feet. Day 
after day, as the diamond seekers from Cape Town plodded on 
with their creaking wagons, the same purpled brown face was 
outspread before them of the stunted flowering shrub which has 
given its name to the desert, spotted with patches of sun-cracked 
clay or hot red sand. To some of the Scotchmen this scrub had 
the cheery face of the heather of their own Highlands, and home- 
sick Englishmen would ramble far through the furze to pick the 
bright yellow flowers of plants that recalled the gorse of their 
island homes. 1 These common bushes, rarely rising a foot in 
height, and the thick, stunted camelthorn, were almost the only 
vegetable coating of the desert. 

Straggling over this plane ran the quaint ranges of flat- 
topped hummocks and pointed spitz-kopjes, streaked with 
ragged ravines torn by the floods, but utterly parched for most 
of the year. Shy meerkats, Cynictis penicillata, weasel-like crea- 

1 Special correspondence London Chronicle and other English journals, Novem- 
ber, 1 899. 



132 THE DIAMOND MINES OF SOUTH AFRICA 

tures with furry coats, peered cautiously from their burrows at 
the strange procession of fortune-hunters, and from myriads of 
the mammoth ant-hills that dot the face of the desert innumer- 
able legions of ants swarmed on the sand along the track of the 
wagons. Sometimes at nightfall the queer aard-vark lurked 
upon the ant-heap and licked up the crawling insects by thou- 
sands. Far over the heads of the travellers soared the preda- 
tory eagles and swooping hawks, harrying the pigeons and dwarf 
doves that clustered at daybreak to drink at the edge of every 
stagnant pool. 1 

Even in the earliest years of the Dutch advance into South 
Africa, when wild beasts browsed in troops on every grassy plain 
and valley and the poorest marksman could kill game almost at 
will, the karoo was shunned by almost every living creature 
except in the fickle season of rainfall. The lion skirted the 
desert edge warily, unwilling to venture far from a certain water- 
brook or pool. There was nothing on the bare karoo to tempt 
the rhinoceros from his bed in green-leaved thickets, and only 
the wide-roaming antelopes (trekbok) rambled for pasturage far 
over the sparsely coated and parched desert waste. If this was 
true in the days when the tip of Africa was swarming with animal 
life, it is not surprising that the diamond seekers in 1 869 and 1 870 
rarely saw any living mark for their rifles when they journeyed 
over the desert. Rock-rabbits, akin to the scriptural coney, 
scampering to their holes, were often the largest game in sight 
for days at a time, and it was counted remarkable luck when any 
hunter put a bullet through a little brown antelope, a grysbok, or 
springbok. 2 The springboks still haunted the Great Karoo, for 
they were particularly fond of its stunted bush growth, and in 
the rainy season many droves of these antelopes could be seen 
browsing warily or flying in panic from the spring of the cheetah, 
the African hunting leopard ; but most of the bigger game, bles- 
bok, haartebeest, koodoo, and wildebeest, that used to feed 

1 "A Breath from the Veld," John Guille Millais, London, 1895. "Among 
the Diamonds," 1870-1871. 

2 "The Diamond Diggings of South Africa," Payton, 1872. 



THE DISCOVERY 133 

greedily on the same pasture, had been killed or driven away by 
the keen hunting of the years that followed the taking of the 
Cape by the English.' 

Sometimes the clear sky of the horizon was blurred by the 
advancing of monstrous swarms of locusts, the " black snow- 
storms " of the natives, sweeping over the face of the land like 
the scourge of devouring flames, chased by myriads of locust 
birds, and coating the ground for miles around at nightfall with 
a crawling, heaving coverlet. Then might be heard the hoarse 
trump of the cranes winging their way over the desert and drop- 
ping on the field strewn with locusts to gorge on their insect 
prey. Or the travellers saw the slate-white secretary bird stalk- 
ing about with his self-satisfied strut and scraping up mouthfuls 
with his long horny bill. 

More marvellous than the locust clouds were the amazing 
mirages that deceived even the keen-eyed ostriches with their 
counterfeit lakes and wood-fringed streams, so temptingly near, 
but so provokingly receding, like the fruits hanging over the 
thirsting Tantalus. Sometimes hilltops were reared high above 
the horizon, distorted to mountainous size and melting suddenly 
in thin air or a flying blur. Now a solitary horseman was seen 
to swoop over the desert in the form of a mammoth bird, or a 
troop of antelopes were changed to charging cavalry. No trick 
of illusion and transformation was beyond the conjuring power 
of the flickering atmosphere charged with the radiating heat of 
the desert. 2 

When the prospectors crossed the karoo and entered the 

1 " A Breath from the Veld," John Guille Millais, London, 1895. 

2 Despatches of Julian Ralph and other special correspondents to London jour- 
nals, October-December, 1899. "Sketches and Studies in South Africa," W. J. 
K. Little, London, 1899. " Portraits of the Game and Wild Animals of Southern 
Africa," W. G. Harris, London, 1840. "The Large Game and Natural His- 
tory of South and Southeast Africa," W. H. Drummond, Edinburgh, 1875. 
"Travel and Adventure in Southeast Africa," F. C. Selous, London, 1893. 
"Kloof and Karoo," H. A. Bryden, London, 1889. "Days and Nights by 
the Desert," P. Gillmore, London, 1888. "Gun and Camera in South Africa," 
H. A. Bryden, London, 1893. 



134 THE DIAMOND MINES OF SOUTH AFRICA 

stretches of pasture land which the Dutch called veld, the scenes 
of their marches were much more lively and cheery. Little farm- 
houses dotted the plains and valleys, rude cottages of clay-plas- 
tered stones or rough timbers, but hospitable with fires blazing 
on open hearths, big iron pots hanging from cranes and simmer- 
ing with stews, and broad-faced, beaming vrouws and clusters of 
chunky boys and girls greeted the arrival of an ox-wagon from 
the coast as a welcome splash in the stagnant stream of their daily 
life. 1 

At some of the halting places on the banks of streams, or 
where plentiful water was stored in natural pans or artificial 
ponds, the extraordinary fertility of the irrigated soil of South 
Africa was plainly to be seen in luxuriant gardens, with brill- 
iant flower-beds and heavy-laden fruit trees and vines. Here 
figs, pomegranates, oranges, lemons, and grapes ripened side by 
side, and hung more tempting than apples of Eden in the sight 
of the thirsting, sunburnt, dust-choked men who had plodded 
so far over the parched karoos. They stretched their cramped 
legs and aching backs in the grateful shade of spreading branches, 
and watched with half-shut eyes the white flocks nibbling on the 
pasture land, and the black and red cattle scattered as far as the 
eye could see over the veld. Tame ostriches stalked fearlessly 
about them, often clustering like hens at the door of the farm- 
house to pick up a mess of grain or meal, apparently heedless of 
any approach, but always alert and likely to resent any familiarity 
from a stranger with a kick as sharp and staggering as any ever 
dealt by a mule's hind leg. 

The interior of the homes in these oases was. not so invit- 
ing, for the rooms, at best, were small and bare to the eye of 
a townsman. But some were comparatively neatly kept, with 
smoothly cemented floors, cupboards of quaintly figured china 
and earthenware, hangings and rugs of leopard, fox, jackal, and 
antelope skins and brackets of curving horns loaded with hunt- 
ing arms and garnished with ostrich feathers. For the guests 

1 " Among the Diamonds," 1870-1871. "The Diamond Diggings of South 
Africa," Payton, 1872. "South Africa Diamond Fields," Morton, 1876. 



THE DISCOVERY 135 

there was probably the offer of a freshly killed antelope or sheep ; 
but the farmer's family was often content with " biltong," the 
dried meat that hung in strips or was piled in stacks under his 
curing shed. 

Near every house was the accompanying kraal or open-walled 
circle for the confinement of the flocks at night, built of stones, 
and usually so bedded and filthy with fresh dung that a heavy 
percentage of the farmers' sheep died yearly from foot-rot or scab. 1 
Close to the kraal was the water reservoir for the flocks and the 
household use, unless the farm lay on the bank of an unfailing 
stream. These collections of water were commonly hill drainage, 
stored in long, narrow ponds by rough dams across ravines, or 
the drainage and rainfall filling shallow natural basins which the 
Boers call "pans." In the early morning the birds flew from ail 
quarters to these ponds. Wild ducks, geese, plover, sandgrouse, 
and flocks of pigeons and doves hovered over the pools and 
splashed and dabbled in the water, while the blue-gray Kafir 
cranes stalked warily along the brink. 

These basins are quite numerous in the country lying 
between the Orange and the Vaal, as well as throughout the 
Transvaal. The light earth washed down the hill slopes was 
largely calcareous, and incrusted the grasses and roots of the 
basin in a calc-tufa which is almost impervious to water. So 
the pans became excellent natural reservoirs, though there was, 
of course, a heavy loss from evaporation. No calamity is so 
dreaded by the graziers as the failure of their water-supply, for 
it has often caused the loss of a flock and the ruin of the poor 
owner. Therefore the pans are highly valued and strictly re- 
served, and the dams are daily inspected lest a burrowing land 
crab should open the way for a rush of water that would empty 
the reservoir. 2 When a settler was fortunate in getting a tract 
of land with a pan or a water-spring, he almost invariably gave 
the name to his farm, as Dutoitspan, Dorstfontein, Jagersfontein, 

1 " On Veld and Farm," Frances MacNab, London, 1897. "South Africa 
Diamond Fields," Morton, New York, 1877. 

2 " Among the Diamonds," 1870-1871. 



136 THE DIAMOND MINES OF SOUTH AFRICA 

Bultfontein, names of inconsiderable little patches on the face 
of South Africa, which were destined to become memorable by 
approaching revelations. 1 

Attracted by the good pasturage and water and the sight of 
flowers, fruits, and birds, even the eager diamond seekers were 
not loath to linger for a day at one of these oases and rest them- 
selves and their cattle before pushing on to the Vaal. As they 
drew near to their goal the face of the country began to change. 
After passing the Modder River, the grassy plains stretched 
out wider and longer and more gently undulating, and the mirage 
was more greatly magnifying and illusive. Herds of wild game, 
chiefly springbok, blesbok, hartebeest, wildebeest, and koodoo, 
were now frequently seen, and the ears of the travellers were 
tickled with the cheery karack-karack of flying korhaan and 
the pipes of red-legged plover. There are black headed or veld 
korhaan and bush korhaan. These birds, which are very plenti- 
ful along the Vaal River and about Kimberley, belong to the 
smaller bustard species. The cock bird of the veld korhaan has 
a black head with white spots on the sides. The top of the 
head or crest is of a reddish gray color. The back is also red- 
dish gray, the markings of the feathers being in rings or stripes. 
The wings are black-and-white, and the legs yellow. The hen 
birds have reddish gray heads, but otherwise are similar in feather 
to the cock bird. The bird derives its name from the Dutch 
word knor, to scold, and haan, hen or bird, on account of the 
scolding noise made by the male bird as it rises from the ground. 
The original word, knorhaan, has been corrupted into korhaan. 
The bush korhaan has a gray head with a light blue patch on the 
crown, just back of which is a pink-brown crest an inch and a 
half long. The back is covered with brown-and-white feathers 
with diamond-pointed markings. The lower part of the leg is 
yellow and the upper part blue. The Dutch call one variety 

1 " Achtzehn Jahre in Sud Africa," E. J. Karrstrom, Leipzig, 1899. " Seven 
Years in South Africa," Emil Holub, London, 1881. "South Africa," A. H. 
Keane, London, 1895. "South Africa of To-day," Captain F. E. Younghusband, 
London, 1898. "Ten Years in South Africa," J. W. D. Moodie, London, 
1835. "South Africa," George McCall Theal, 1888-1893. 



THE DISCOVERY 137 

of birds somewhat resembling the bush korhaan rudely " dik- 
kops," thick heads, from their appearance when wounded ; but 
they are none the less handsome birds, and they were eagerly 
shot and eaten by the diamond seekers on the way to the fields 
and in the camps on the river. There were great numbers, too, 
of the paauw or cape bustard near the Modder River, and red- 
winged partridges and Guinea fowl that gave a welcome variety 
to the meals of the travellers. 1 

Over the rolling ground the prospectors pressed rapidly to 
the Diamond Fields and soon reached the river border where the 
plains ran into the barrier of ridges of volcanic rocks. Jolting 
heavily over these rough heaps and sinking deeply in the red 
sand wash of the valleys, the heavy ox-wagons were slowly tugged 
to the top of the last ridge above Pniel, opposite the opened 
diamond beds of Klip-drift, where the anticipated Golconda was 
full in sight. Here the Vaal River winds with a gently flowing 
stream, two hundred yards or more in width, through a steeply 
shelving, oblong basin something over a mile and a half in length 
and a mile across. A thin line of willows and cotton-woods 
marked the edge of the stream on both banks. On the descend- 
ing slope toward the river stood the clustering tents and wagons 
of the pilgrims waiting to cross the stream. 

In the dry season the Vaal was easily fordable by ox-wagons 
at a point in this basin, and the ford, which the Boers call 
" drift," gave the name to the shore and camp opposite Pniel, 
"Klip-drift," "Rocky-ford." When the river was swollen 
by rains, the impatient fortune-hunters were forced to wait, fum- 
ing, in sight of the diamond diggings until the flood subsided ; 
but, a few months after the rush began, a big, flat-bottomed ferry- 
boat, called a punt, was constructed to carry over the wagons 
and cattle, while the men crossed in rowboats, making regular 
ferry trips between Pniel and Klip-drift. 

How stirring were the sights and sounds from the ridge at 
Pniel to every newcomer while the swarming diamond seekers 
were crossing the river and spreading out over the northern 
1 "Among the Diamonds," 1870-1871. 



138 THE DIAMOND MINES OF SOUTH AFRICA 

bank! the confused clustering at the ford the rambling of 
stragglers along the shore the gravel cracking and grinding 
under the hoofs of the horses and ponies racing along the bank 
and rearing, plunging, and bucking at the check of the bits and 
prick of the spurs the outspanning and inspanning of hun- 
dreds of oxen the swaying and creaking wagons the writh- 
ing, darting lash of the cracking whips of the drivers the 
sulking, balking oxen, driven into long, straining lines that 
dragged the ponderous, canvas-arched "prairie-schooners" 
through the turbid water and over the quaking sands the 
whistling, shouting, yelling, snorting, neighing, braying, squeak- 
ing, grinding, splashing babel the scrambling up the steep 
Klip-drift bank the scattering of the newcomers the 
perching of the white-topped wagons and the camp-tents like 
monstrous gulls on every tenable lodging place on bank, gully, 
and hillside the scurrying about for wood and water the 
crackling, smoking, flaming heaps of the camp fires the 
steaming pots and kettles swinging on cranes the great placer 
face, pockmarked with holes and heaps of reddish sand, clay, 
and gravel the long stretches of the miners' rockers and 
troughs at the water's edge and chief of all in interest, the 
busy workmen, sinking pits and throwing out shovelfuls of 
earth, filling buckets and hauling them up with ropes, loading 
and shaking the rockers, driving carts full of heavy gravel to 
the water troughs, returning for new loads, scraping and sorting 
the fine, heavy pebbles on tables or flat rocks or boards spread 
on the ground ! 

No labored, crawling recital can compass and picture in 
print any approach to the instant impress on the eye and ear 
of the moving drama on the banks of the Vaal. Observer after 
observer groped vainly for graphic comparison. " Klip-drift is a 
swarm of bees whose hive is upset," said one , " a bank lined 
with ant-hills," wrote another, prosily ; " a wild rabbit warren, 
scurried by a fox," ventured a third ; " an insane asylum 
turned loose on a beach," sneered a fourth. It was a mush- 
room growth of a seething placer-mining camp in the heart 



THE DISCOVERY 



'39 



of the pasture lands of South Africa. To old Australian and 
American miners it had a patent likeness to familiar camps and 
diggings, but its local coloring was glaringly vivid and unique. 1 

1 " Among the Diamonds," 18701871. " The Diamond Diggings of South 
Africa," Charles Alfred Payton, London, 1872. " South Africa Diamond Fields," 
Morton, New York, 1877. "To the Cape for Diamonds," Frederick Boyle, 
London, 1873. "Diamond Fields of South Africa, by One who has visited the 
Fields," New York, 1872. 




Pniel Diggings. 



CHAPTER V 




THE CAMPS ON THE VAAL 

EFORE calling to view the spreading of the 
diamond seekers along the line of the Vaal 
River, the rearing of successive camps, and the 
growing pursuit of gems in the gravel, it is 
essential to trace the progress of diamond 
mining from its original development on the 
water-shed of the Indus, and to account in great measure for 
the blundering, confusion, and failures in the new Diamond 
Fields by showing how crude and imperfect were any known 
methods of winning the precious stones at the time of the South 
African discoveries. 

From earliest history there had been no change and no 
prospect of change in the diamond mining of India (described 
in Chapter i). In the Deccan diamond fields, as in the other 
congested districts, there was such an influx of poor natives that 
no labor-saving contrivances were sought for, and the diamond- 
bearing gravels were lifted and washed by hand as they had been 
by the first generation of workers. There had been no compe- 
tition with the Deccan field, and no considerable production 
outside of it, until the diamonds of the Brazilian fields were 
made known to the Portuguese in the year 1728. As soon as 
the Home Government learned of this discovery, the diamonds 
in Brazil were declared to be State property, and for a hundred 
years diamond mining was a Crown monopoly. This con- 
dition was a clog to any possible advance in the methods of 
mining. There was a constant drain on the industry without 
any effort to develop it systematically, thoroughly, or economi- 
cally. 

140 



THE CAMPS ON THE VAAL 141 

The chief deposits were found, at first, in river beds and 
ravines in a breccia of clay, quartz pebbles, and sand, charged 
with oxide of iron. Some of the richest beds were opened along 
the rivers Jequetinhonha and Pardo in the valley of Sejues, and 
on the line of the rivers Aboite, Andaja, da Serreno, da Prata, 
and San Francisco in the province of Minas Geraes. 1 

The diamond-bearing ground was worked under govern- 
ment agents or leased to contractors. Quick returns were the 
first object. So gangs of slaves were put on the grounds, 
regardless of loss, if only the cream of the fields was skimmed. 
In the dry season the beds of the smaller sierran streams were 
nearly or wholly dry. Underlying the surface wash of sand in 
the bed was the formacao or cascalho, heavy diamond-bearing 
gravel intermixed with boulders. The alluvial soil was gen- 
erally from eight to twenty feet thick, a silicious sand chiefly, 
deep colored by ferruginous clay. The diamonds and other 
minerals of high specific gravity were held in the bottom layer 
of this alluvium, usually cemented in a coarse pudding-stone of 
quartz and itacolumite the cascalho. The sand was rudely 
scraped away or carried off in pans, the boulders pried out, and 
the cascalho exposed. Then the gravel was collected labori- 
ously in pans and piled in heaps to await the rainy season, when 
the streams filled the dry courses and there was water at hand 
for washing the gravel. 

Bacus or shallow pits were sunk in the sand along the brink 
of the streams, and in these pits a few panfuls of gravel were 
thrown. The bottom of the bacu was made to slope so that 
the dashing of water on the gravel heap would readily wash away 
the clinging sand and the lighter and larger stones. The expert 
slaves washed the heaps in the bacus with splashes of water cast 

1 "The Diamond Fields of Brazil," Report of United States Minister Bryan, 
March i 2, 1 899, conveying report of American Secretary of Legation, Dawson. 
"A Treatise on Gems," Lewis Feuchtwanger, M.D., 1867. "An Account 
of Diamonds found in Brazil," James Castro de Sarmente, M.D. "Genuine 
Account of the Present State of the Diamond Trade in the Dominions of Portugal," 
a Lisbon merchant, London, 1785. "Travels in South America," J. J. von 
Tschudi. 



142 THE DIAMOND MINES OF SOUTH AFRICA 

from concave wooden plates with a peculiar whirl which has- 
tened the separation of the heavier gravel. This concentrate, 
containing most of the diamonds in the cascalho, was then 
washed again in a batea, a wooden dish with a depression in the 
centre. By dexterous shaking and whirling motions of the batea 
filled with water and a few handfuls of gravel, the lighter gravel 




Delports' Hope, Vaal River Diggings. 

was carried to the rim and washed or scraped away, and dia- 
monds mixed with heavier pebbles were collected in the hol- 
lowed centre of the dish. A gentle tilt of the batea drained off 
the water, and the precious stones were picked from the other 
pebbles by hand. 

Sometimes the formacao was deposited in an inclined mov- 
able trough or cradle on whose face fifteen to eighteen pounds 
were spread out at a time. Then a carefully regulated stream 
of water was allowed to run through this deposit into a lower 
trough and gutter while the cradle was rocked continually. 
When the water ran off clear from the lower trough, the work- 
ing negro would pick out the stones in the cradle with his fin- 
gers, until only the finest pebbles remained, which he scraped 
over and examined with the closest attention to detect the pos- 
sible presence of diamond crystals. 1 

1 "A Treatise on Gems," Feuchtwanger, 1867. Report of United States 
Minister to Brazil, March, 1 899. 



THE CAMPS ON THE VAAL 



This was a slow and tedious process, at best. The percent- 
age of precious stones won from the gravel necessarily depended 
on the care, expertness, and eyesight of the workers. Experi- 
ence proved that fairly expert gold placer miners were not 
equally competent in handling diamond-bearing gravel, and slave 
labor was not diligent or trustworthy. The loss was increased 
by the greedy pressure for big and quick returns, and the pre- 
mium set on the extraction of large stones. 

When, in the course of mining, streams were diverted from 
their beds by dams and sluiceways, there was urgent need of 
hurrying, for the frail dams could not bear the rush of a flood 
in the rainy season, and it was necessary to remove the gravel 
from the stretches of river beds before the heavy rains fell. 



_""**--. _ , ^ 

^S : ^x."*- " *" 

^S3&^* * 

:r^b^^fe 




-'* .;>" ' J Irw*?. 

Diggers' Camps on the Vaal River. 

Often the formacao was buried under thirty feet or more of 
sand, and all this overlying mass had to be scooped up and 
carried off as well as the layer of gravel. As the slaves had 
nothing better than pans for this work, the beds were covered 
with swarms of negroes bearing pans on their heads and nibbling 
away at the ground like ants in the effort to reach the gravel 
before the floods came. In the reckless haste many tracts of 



144 TH E DIAMOND MINES OF SOUTH AFRICA 

diamond-bearing gravel were buried under ground too deep for 
profitable working, or covered by the waste of flooded rivers. 

As the mines advanced up the hillsides, following the course 
of the mountain streams, it was seen that there were gupiaras or 
deposits of diamond-bearing gravel along the steep slopes of the 
ravines, and these were worked by carrying the gravel to the 
banks of streams, or by cutting sluiceways to the deposits. 
Finally, on the sierran ridges and plateaus the conglomerate beds 
were reached, from which the deposits in the river beds had 
been washed by the mountain streams. This conglomerate was 
chiefly itacolumite, 1 a micaceous sandstone, accompanied by 
mica-schist and penetrated irregularly by quartz veins. This 
was the prevailing composite in the Serro de San Antonio, in 
which the Jequetinhonha rises in the Serro de Matta de Corda, 
the fountain head of the Rio Francisco. 1 

Here the diamonds were not as thickly sprinkled as they 
were in the cascalho concentrate, but the quantity was sufficient 
to make extraction profitable, if the conglomerate could be dis- 
integrated and washed. This was effected by collecting rain 
water in pools at points above the conglomerate and carrying 
down the water through ditches into gullies cut in the beds. 
By the flow of the water, the formacao was separated from the 
mass of rocks and sand. This device worked well, but owing 
to the scarcity of water, the washing could only be continued for 
a few weeks, at most, in the course of a year. In 1832 mining 
in these fields was opened to the public, but the most accessible 
and prolific beds had been worked, and there was little apparent 
encouragement for the investment of capital in any large under- 
taking which might have advanced the science of diamond 
winning. It is said that more than half of the diamonds pro- 
duced in Brazil were stolen by the workmen and sold to contra- 
band dealers, by whom they were secretly sent out of the country. 

Outside of the Indian and Brazilian fields no considerable 
source of supply had been discovered anywhere. Some dia- 
mond-bearing ground had been found in Borneo, which yielded 
for many years a dribbling return, and in 1829 the first-known 



THE CAMPS ON THE VAAL 145 

diamond of Russia was discovered on the west flank of the Ural 
Mountains by Humboldt and Rose, in a gold placer field near 
the iron mines of Bissersk. Here the prevailing rock forma- 
tion, like that in the upper diamond fields of Brazil, was itacolu- 
mite, with an admixture of mica and iron pyrites. 1 The debris 
washed into a few valleys beneath this range yielded a meagre 
return to the searchers, but there was nothing to inspire any 
ardent working, and in Bohemia, Australia, Mexico, and the 
United States, the picking up of a few isolated specimens was 
noted as a curious occurrence rather than as the foundation of 
any hope of a productive diamond field. 2 

So, at the time of the discoveries of diamonds on the banks 
of the Vaal River, there was no known method for the extraction 
of diamonds beyond the shovel of the Indian, the batea of the 
Brazilian, or the cradle of the gold miner. There was no antici- 
pation, on the part of the diamond seekers, of any formation in 
Africa except the diamond-bearing gravel of alluvial deposits, 
and the prospectors of the first rush did not seek for diamonds 
beyond the gravel along the banks of the Vaal. 

The Early Mining at Klip-drift, now called Barkly West. 

The first waves of the influx from the southern country and 
coast towns were warmly greeted by the small parties at work 
on the Vaal. The diggers were squatters, without any legal 
title to an inch of the river bank, as they very well knew. But 
they relied on actual possession without contest, for their rocky 
field was so apparently worthless that no farmer had cared to 
secure it. They did not trouble their heads with any question- 
ing whether the South African Republic covered their shore 
line, or whether any native tribe laid claim to it, but they were 

1 " A Treatise on Gems," Feuchtwanger, 1867. "Notices sur les Diamants 
de 1'Oural," Parrot. "Transactions of the Imperial Russian Mineralogical Soci- 
ety," at St. Petersburg, 1842. " De Novis quibusdam Fossilibus quae in mon- 
tibus Uraliis inveniuntur," Gustav Rose, 1839. 

2 "Gems and Precious Stones of North America," Kunz, 1890. 



146 THE DIAMOND MINES OE SOUTH AFRICA 

so weak in numbers that they had some fear of possible attack 
from the neighboring Koranas and Griquas, or other natives 
who might covet their oxen and arms and supplies, as well as 
their hard-won gems. 1 In view of the abject state of the few 
surviving Hottentots on the Vaal, any dread of their hostility 
seemed absurd, but the miners did not know how weak the 
natives were, and their new-found treasure unsteadied their 
nerves. So they were glad to see a rally of prospectors on the 
fields large enough to scare off any menacing natives. 

The. early comers picked out irregular patches of ground 
here and there, to suit their fancy, and dug and strayed along 
the river banks as they pleased, prospecting on any unoccupied 
spot. There was no precise limit to the size of any claim. One 
party would pounce on a whole hillock, like the prolific " Natal 
kopje," and another would occupy a hundred feet or more of 
shore line. There was no apparent need of jostling one another, 
when any square rod for miles along a river bank was as thickly 
sprinkled with diamonds as another, so far as any of the pro- 
spectors could judge. Still, the known yield of the Natal kopje 
drew preference to locations around it, and the product of other 
neighboring placers was so enticing that the mass of diggers 
concentrated at Klip-drift. 

This massing made it necessary to agree on some defined 
limits of ground which a man could reserve for his own work- 
ing, or combine with the sections assigned to companions. To 
fix and make this assignment a " Diggers' Committee " was 
chosen by an informal mass meeting of the prospectors, which 
made simple regulations controlling the working of the river 
diggings. It was agreed that the size of a location should be 
thirty feet square, and that title should be conveyed by a certificate 
from the supervising committee. The water's edge along the 
river was open to anybody wherever it was possible to set a 
trough or a miner's cradle without interfering with other ground- 
washing fixtures already in place, but locations might begin a few 
yards from the river. 2 So there was soon a close-set fringe of 

1 "Among the Diamonds," 1870-1871, John Noble. 2 Ibid. 



THE CAMPS ON THE VAAL 

cradles and water-troughs at the bottom of the Klip-drift bank, 
and the ridged and gullied slope for hundreds of yards inland 
was pitted with holes from ten to thirty feet square, and ranging 
in depth from four to twenty-five feet. If the river shore in 
line with the parallel claim was too thick set with cradles to 
admit a new washing machine, or if the claim was high up on 
the bank, water for washing was sometimes carried up from the 
river in carts to the working ground. Alluvial soil covered the 
face of the basin, more or less thickly, for a stretch of half a 
mile from the river, lying even on the tops of the kopjes, except 




Diggings at Gong Gong, 1880. 



where rugged boulders and blocks of basalt and trap protruded 
stiffly above the coating of gravel. 

The choice of location was largely determined by fancy, 
rather than any solid reason. Some preferred light colored 
patches of gravel to dark, but would have been puzzled by any 
call to justify their choice. Others sought for tops of kopjes, 
with a supposition that the rains had washed the light gravel 
downhill and left the heavier deposit with the diamonds on the 
crown of the hillocks and ridges. 1 It was generally observed, 
however, that diamond crystals were most plentiful in spots 

1 "The Diamond Diggings of South Africa," Charles Alfred Payton, London, 
1872. 



148 THE DIAMOND MINES OF SOUTH AFRICA 

where garnets and peridot were thickly deposited in the gravel, 
and this observation was in accord with current accounts of 
mining in other diamond fields. So the occurrence of these 
red and green pebbles was commonly hailed as an assurance of 
the presence of diamonds, and gravel so charged was washed 
and sorted with exceptional care. But there was no concentrated 
deposit in this field like the cascalho in the Brazilian river 
valleys, and the labor of washing the thick mass of loose gravel 
was necessarily great. 

There were no appliances for handling and concentrating the 
gravel marking any noticeable advance above the slow and labo- 
rious methods of the Brazilian and Indian placer workers. The 
deposit was a mass of gravel and sand, thickly sprinkled through- 
out with heavy boulders of basalt and melaphyre which were 
laboriously pried and dragged out of the shallow pits sunk by 
the miners. 1 The mixed gravel and sand was shovelled into 
wheelbarrows or carts and taken to the river's edge, where it was 
dumped into heaps on the ground, or in troughs sunk in the 
bank. Then the gravel was washed in cradles, with two or 
three screens of perforated iron, or zinc, or wire mesh, set to 
form partitions with discharge holes so graduated that the larger 
stones were held above the upper and coarser screen, while the 
sand and lighter gravel flowed out through the upper and lower 
screen holes. Meanwhile the cradle was more or less expertly 
shaken to cause a deposit of the gravel of high specific gravity 
on the bottom between the screens. The worthless stones in 
the upper part of the cradle were then picked and scooped out 
by hand and thrown away, while the concentrate was taken out 
carefully and carried to the sorting table, an ordinary deal stand, 
or any level wooden or iron structure, or to a flat stone. Here 
the deposit was spread out thinly and sorted over inch by 
inch with a short scraper of hoop iron, or any other thin strip, 
while the appearance of a diamond was more or less keenly 
watched for. 2 

1 "The Diamond Diggings of South Africa," Charles Alfred Payton, London, 
1872. 2 Ibid. 



THE CAMPS ON THE VAAL 149 




Vaal River Diggings. 

This washing machine was practically the same as the Aus- 
tralian gold placer miner's cradle, or the American rocker, and it 
had been used for years on the Brazilian diamond fields, though 
the screening of the Vaal was probably more exact. But the 
Brazilian negroes had become far more expert by long practice 
and training than the green workers on the line of the Vaal, and 
the handling of the concentrate in their bateas was extraordinarily 
deft. It has been demonstrated over and over again in placer 
fields that inexperienced washers cannot compete with trained 
hands in concentrating gold dust, and even expert gold placer 
workers often failed to handle diamond-bearing gravel efficiently. 
So it is not surprising that many of the awkward adventurers 
in the new fields lost heart completely at their failure to extract 
any diamond from the masses of gravel which they dug and 
washed so laboriously ; and it is practically certain that the per- 
centage of gems saved, at first, was below the average winning 
from the Brazilian sands. 

The irregularity of the distribution of diamonds in the shore 
bed was greatly perplexing and disappointing to the groping 
locaters. The precious stones were strewed in the gravel in a 
scattering way that defied any calculation. Here and there was 
a rich patch of ground, while tracts all around it, precisely simi- 
lar in a surface view, held only a few small diamonds or were 



150 THE DIAMOND MINES OF SOUTH AFRICA 

hopelessly barren. Even in the best placers there were apparent 
freaks of deposit that sorely puzzled the diggers, and almost pro- 
voked the belief in the dropping of the gems by whimsical genii 
rather than by the play of natural agencies. One man, working 
side by side with another for weeks in adjoining claims, would 
not find one precious stone, while his neighbor was adding daily 
to his little sparkling heap. Even when claims were so split up 
that a digger could hardly turn about without brushing against 
a comrade there was the like insolvable contrast of gem-studded 
gravel and worthless pebbles. Often, too, when a claim had 
been abandoned by an unlucky miner, the next man who jumped 
into the deserted hole would unearth in a day a superb diamond, 
and, perhaps, wash out in a week a score more of precious 



stones. 1 



The miners were, as a body, so orderly, so tenacious of their 
own rights under the established regulations, and so prudent in 
restricting the possible extent of monopolized ground, that there 
was little " claim jumping " or bitter wrangling. The provision 
against loafing or the holding of unworked claims on speculation 
was sufficiently sharp. The neglect to work a claim for three 
days consecutively forfeited the holder's license, and the ground 
was then open for the issue of a new certificate to the first claim- 
ant. For many months all unoccupied ground in the Klip-drift 
camp was greedily pounced upon by newcomers to the fields. 
So this part of the river basin was continuously covered with a 
busy swarm of workers, digging, washing, sorting, driving carts, 
and stirring in all the daily occupations of camp life. Where 
one man lost heart and went off prospecting up or down the 
river, or plodded wearily homewards, another was ready to take 
his place in a moment and continue the unflagging round of 
work. 

It was soon perceived that such diamond placer digging 
was inevitably a gambling speculation, and few complained 
loudly of their hard luck, or bitterly grudged the success of 
their neighbors. When an unusually large stone was found, 

1 "Among the Diamonds," 1870-1871, John Noble. 



THE CAMPS ON THE VAAL 151 

there was commonly a shout and a rallying of exultant friends 
around the lucky finder, and all through the fields a redoubled 
fervor of work from the spur of the signal success. Every one 
felt that the good fortune of a comrade might be his own the 
next moment, and,if this hope was cast down, the diggers toiled 
on with indomitable pluck and sanguine spirit, ever lifting the 
glittering image of better luck some day. So the rasping of 
shovels, the splashing of gravel, the rumbling of carts, the 
dumping of loads, and the rattle of cradles went on incessantly 
with a lively din from morning till night. 




River Diggings. Waldek Plant. 

For the sorting of the concentrated gravel shady spots were 
chosen beneath spreading tree-branches, where tables were set, 
or under the cover of canvas screens stretched over posts. 
Here the miners bent over the thin layer of gravel, scraping 
along the pebbles bit by bit, and gluing their eyes to the sliding 
stones in anxious search for the coveted tiny white crystals ; or 
stretched out at full length on their stomachs, they scraped the 
gravel over the face of the boards or iron sheets laid flat on the 
ground. In this branch of diamond winning, where keen eyes 
were essential, the native blacks were largely employed, some- 
times under close watch of a white overseer, and sometimes 
without any oversight. Part of the black sorters were strictly 



152 THE DIAMOND MINES OF SOUTH AFRICA 




Pniel Diggings, Vaal River. 

faithful and honest, as was shown by test after test. One 
boy brought straight to his master a diamond of eighty carats, 
which his quick eye detected in the roots of an old stump that 
he had been told to dump into the river. Another returned the 
counterfeit stones that his employer had purposely dropped in 
the concentrate. 1 But all were not equally trustworthy, and 
many fine stones were filched from the tables by nimble-fingered 
sorters, even under the eye of a wary overseer. When the Boer 
farmers came to the fields, they often brought their families with 
them, and it was a common sight to see father and sons digging 
and washing, while the mother and daughters sat on the ground 
industriously picking over a layer of pebbles. Sometimes, too 




Klip-drift, Early River Diggings. 

1 "Among the Diamonds," 1870-1871, John Noble. " The Diamond Dig- 
gings of South Africa," Payton, 1872. 



THE CAMPS ON THE VAAL 



'53 



the wives and sisters of Knglish miners, even women who had 
rarely soiled their white hands before, might be seen sorting 
river gravel as ardently as any prospector on the line of the 
Vaal. 

When newcomers roamed about sight-seeing over the fields, 
they were surprised to note how rarely their presence drew even 
a fleeting glance. Scarcely any one of all the groping swarm of 
diggers, washers, and sorters, white or black, men or women, 
diverted an eye for a moment from the intent absorption of the 
search for the tiny crystals embedded in the vast stretches of 
gravel. Eternal vigilance is the watchword of diamond winning 
as well as of liber- 
ty. It was keenly 
felt by the dia- 
mond seekers that 
a fortune might 
slip through their 
hands in the shift- 
ing and twinkling 
of an eye. So wan- 
dering strangers 
threaded their 
way among the 
burrows in the Gong Gong. 

pitted bank and the diamond sorting tables without attracting any 
more attention than stray pebbles rolling down the gravel heap. 1 
Whenever any one of this curious swarm found a big stone 
he had a prize in his hands, for the precious crystals of the 
Vaal river beds are exceptionally good and free from fractures. 
There were few stones ranging over thirty carats, but ten carat 
stones were not uncommon, and even the tiniest stones of one 
carat or less were usually well shaped. Some were lightly tinged 
with yellow, detracting somewhat from their market value, but 
there was a large percentage of stones perfectly white, or so nearly 

1 "Among the Diamonds," 18701871. "The Diamond Diggings of South 
Africa," Pay ton, 1872. 




154 THE DIAMOND MINES OF SOUTH AFRICA 

colorless as to defy any scrutiny except that of experts. Deep 
orange yellow stones were occasionally found, and shades of 
yellow grading to the finest straw color were represented as well 
as pale blue, brown, and pink, and other hues ; but any color 
except white or yellow was rarely to be seen. The commonest 
crystalline form was the octahedron, but perfect dodecahedrons 
were not unusual, and twin stones or a conglomeration of crys- 
tals sometimes appeared. There was no adhering film or enve- 
lope such as commonly dulls the lustre of the Brazilian diamond 
crystals. The stones of the Vaal are clear and bright. 1 

Digging for diamonds never becomes dull drudgery, for there 
is always the glittering possibility in the mind's eye of upheaving 
a king's ransom with the turn of a shovel, and it is far more 
exciting to a novice than mining for gold or any other minerals. 
But the diggers on the Vaal River fields soon learned that the 
actual disclosure of a diamond on the face of the gravel which 
he was shovelling was a very rare occurrence, for only the largest 
stones were likely to be seen in a mass of earth and pebbles, and 
few even of these were actually detected in the sinking of the 
pits on the river banks. So the miners were rarely so absorbed 
in their search that they worked without stopping to eat, but 
they clung to the last gleams of the sun as the miners have done 
in the rich gold pocket placers of America and Australia. The 
diggers and washers went to work usually at the same hour, 
about sunrise, took an hour off for breakfast, and for dinner or 
lunch, and stopped work when the sun went down. In the 
hotter weeks of the African summer season (the summer 
November, December, January, and February is the hot as 
well as the wet season) they did little or no work in the midday, 
and when heavy rain and hail storms swept over the fields, all 
sought for cover. 

1 " South African Diamond Fields and Journey to Mines," William Jacob 
Morton, New York, 1877. "The Diamond Diggings of South Africa," Payton, 
1872. "Diamonds and Gold of South Africa," Mitchell, 1888. "To the 
Cape for Diamonds," Frederic Boyle, London, 1873. "Diamond Fields of South 
Africa, by One who has visited the Fields," New York, 1872. 

















s_> 






$ 























THE CAMPS ON THE VAAL 



Camping on the banks of the Vaal was rarely unpleasant to 
any one accustomed to a life in the open country, and even the 
townsmen found little to grumble about. As soon as they 
reached the Diamond Fields, the prospectors looked about for 
good spots on which to lodge their wagons and pitch their tents. 
Some took to the fields small circular or " bell " tents, but the 
greater part preferred a square or oblong " wall " tent, commonly 
ten feet long and eight wide. From a central ridgepole, 
propped at each end, the canvas roof was stretched to side posts 
four feet high, from which flaps hung to the ground. This 
shelter served as a home for two or three men, and a storehouse 
for their stinted outfit. It was not spacious, but even a little 
tent was a welcome change from the cramped bunking in mass 
under a wagon cover, and the airy, clean, canvas chamber was 




Washing Diamond Gravel by Machinery at Gong Gong, 1880. 

much pleasanter than the ordinary farmer's sleeping room, as 
many of the prospectors remarked from experience. Even when 
the campers were obliged, for lack of tents, to sleep in their 
wagons, the big arched wagon did not suffer by comparison with 
any Boer's hut on the veld. The tents were pitched, sometimes 
under the cover of the larger trees lining the river bank, and 
sometimes on sheltered slopes, but the mass at Klip-drift were 
bare to the sun, and exposed to the blast of every storm that 
tore through the valley. 

Often these storms were terrific, opening with the rising of a 



156 THE DIAMOND MINES OF SOUTH AFRICA 

yellow streak above the horizon, and the rapid spreading over 
the blue sky dome of rolling masses of heavy, lurid clouds. 




Lightning at Kimberley. 



Then from the coppery bosom of this pall there came such blaz- 
ing streams of lightning in sheets and contorted shafts, siich 
rending explosions of thunder peals, that the awful flare and 
crash would shake the nerves of hardened men. With this 




Day View, Same Scene. 



appalling discharge there poured from the clouds torrents of 
rain, or a volley of huge hailstones rattling on the canvas roofs 




THE CAMPS ON THE VAAL 157 

and driving man and beast to the nearest shelter. 1 As a safe- 
guard from these electric bolts the miners commonly put iron 
lightning rods alongside of their tent poles and insulated them 
with the necks of glass bottles, but the insecurity of this shield 
was evident in the occasional shattering of a tent and the killing 
or maiming of its occupants. 2 

Except for these storms the climate of the Vaal valley was 
generally agreeable. The winter days were particularly pleasant, 
for the sun soon warmed the air even when the nights were so 
cold that ice formed on the face of water-troughs. In midsum- 
mer the days were often exceedingly hot, the mercury rising as 
high as 100 Fah. in the shade; but the dry air was not nearly 
as enervating as the humid atmosphere of summer days in 
Europe or America, and the lightly clothed miners, avoiding the 
midday glare, suffered little. There was a notable exemption 
from sickness throughout the year, except for diarrhoea and 
dysentery, and fever contracted in summer chiefly from the reck- 
less use of unboiled and unfiltered river water. 8 

Plain food of some kind was plentiful and cheap, especially 
maize meal, commonly called mealie meal, and mutton and game 
were brought into the camp from the neighboring Transvaal and 
Free State farming and pasture lands. There were many wild 
fowls, too, that flocked to the valley of the Vaal, and several 
kinds of food fish abounded in the river, especially one resem- 
bling the voracious English barbel, or the catfish of America, and 
the one which the miners called " yellow fish." The chief lack 
in the food supply was cheap and wholesome vegetables for 
the dearth of these and the excess of meat caused a mild form of 
scurvy to appear in the camp. Fuel for cooking was readily cut 
from the trees along the river bank or from the thickets in the 



ravines. 4 



When the choice locations on the Klip-drift bank were taken, 
the influx, continuously moving to the new Diamond Fields from 
the coast, spread up and down the river, and little camps sprang 

1 The Diamond News, Klip-drift, Nov. 4, 1871. 3 Ibid. 

2 "The Diamond Diggings of South Africa," Payton, 1872. 4 Ibid. 



158 THE DIAMOND MINES OF SOUTH AFRICA 

up at Gong Gong, Union Kopje, Delport's Hope, Forlorn Hope, 
Niekerk's Hope, Blue Jacket, Waldek's Plant, Larkin's Flat, 
and other placer diggings, extending from Hebron twenty miles 
northeast of Klip-drift to Sefonell's, sixty miles west. 1 It has 
been estimated that ten thousand diggers, white and black, were 
stretched along the river in this string of camps, and in roving 
parties of prospectors.' 2 Any possible reckoning of the extent 
of a rush of thousands, which nobody could measure exactly or 
tried to measure, was of course a rough guess, but it seems prob- 
able that this guess was not very far from the fact. Such an 
influx of restless adventurers, pouring along a river line in a 

thinly peopled territory in the heart 
of South Africa, as heedless as a 
locust swarm of any questions of 
state sovereignty, or native tribal 
reservations, or mineral right titles, 
was certain to raise a rumpus, if 
any official authority in South Africa 
undertook to drive them away, or 
exact heavy license fees, or even to 
hold them down under strict laws 

The Largest River Diamond ever found i r j T~M A 

in South Africa. Weight, 3303 Car- rigorously enforced. The Austra- 

ats; Value, 3,500. .; an go j d fie j dg J^J f urn i s hed Some 

highly significant object lessons enforcing this certainty, but the 
little Boer Republics were not disposed to learn any lesson 
from the experience of English Colonies. 

The South African Republic claimed the diamond placer bor- 
der north and west of the Vaal as part of its territory, but it was 
content, at first, with the bare assumption that the diggers on 
the northern and western bank were within the confines of its 
domain, without caring to assert its right of control by any 
marked interference with the free proceedings of the diggers. 
It did not regard the upturning of gravel on its border line as 
any menace of serious intrusion within its territory, and the 

1 "Among the Diamonds," 1870-1871. 

3 " South African Diamond Fields," Morton, 1877. 




THE CAMPS ON THE VAAL 159 

neighboring Boer farmers were generally well pleased with the 
opening of ready markets for their produce. Representatives 
of the Republic were recognized as officers of the law at Hebron, 
but there was little attempt to impress any recognition of its 
authority on the camps farther down the Vaal. 1 

So the miners at Klip-drift went on digging and scraping the 
gravel, under their own simple regulations, month after month, 
until their busy camp burst suddenly into an uproar, when the 
news came in that President Pretorius and the Executive Coun- 
cil of the Transvaal Republic had granted to a firm of three privi- 
leged persons the exclusive right to search for diamonds in the 
territory of the Republic for a term of twenty years from June 22, 
1870, subject to a royalty of six per cent upon the value of all 
diamonds discovered. 2 There were some old Australian placer 
miners on the Vaal River Diamond Fields, and they doubtless 
grinned at the thought of the reception that such a proclamation 
would have met with at Bendigo and Ballarat ; but it was not 
necessary for an adventurer to have had a rearing on any gold 
placer field to fire his spirit to revolt against an edict of dispos- 
session and monopoly. It is idle to debate the question of the 
technical legal right of the administration of the South African 
Republic to make this grant. This may be conceded without 
affecting the countering facts of its gross partiality, inexpediency, 
and practical futility. The whole regular army of the United 
States would have been too small to enforce any such disposition of 
its mineral lands after they had been occupied without protest for 
more than six months by squatting placer miners, and bare com- 
mon sense would have sufficed to inform the administration of 
the little South African Republic that it could not give effect to 
its paper monopoly without a succession of fights that would add 
another " Blood River " to the face of South Africa. 

The instant effect of the grant was a universal uprising and 
mass meeting of the Klip-drift camp, and the declaration of the 
foundation of another free and independent Republic on the Vaal, 

1 "Among the Diamonds," 18701871. 

2 "South Africa," George McCall Theal, 1888-1893. 



160 THE DIAMOND MINES OF SOUTH AFRICA 

of which Theodore Parker, one of the leading adventurers, was 
chosen president. 1 This was, on its face, a proceeding that 
smacked of opera bouffe, but, like Janus, it had another face. 
It was a flaunt of determination to cut off every shred of political 
connection with the South African Republic, and hold possession 
of a slice of rich mining land with a Colony which, at some future 
time, if not immediately, Great Britain might be disposed to 
welcome and incorporate with her imperial cluster on the coast. 
If this hope was not openly avowed at first, it undoubtedly ex- 
isted in the minds of many of the diggers, and no time was lost 
in communicating the situation to her Majesty's High Com- 
missioner at the Cape, Lieutenant General Hay. 

It is, however, unlikely that there was any confident expecta- 
tion of the endurance of the new Republic founded on a gravel 
bank whose precious contents were fast fleeting, but the organi- 
zation was set up as a handy resort, on the spur of the moment, 
to make an imposing show of resistance to the authority of the 
South African Republic, and with the idea of shunning the pen- 
alty of forcibly contesting the execution of the monopoly grant 
within a recognized district of its domain. Whatever legal 
unsoundness there may have been in the construction of the 
Klip-drift Republic, and in the notions of its framers, the shaky 
ship of state served its main purpose. The administration of 
the Transvaal Republic realized their grave blunder too late, 
and being humane and peace-loving men, refrained from any 
attempt to maintain their grant or their contested authority by 
force of arms. But they complained earnestly to the British 
Colonial authorities of the intrusion and illegal occupation and 
insubordination of the squatting adventurers on the Vaal. 

Meanwhile the diamond diggers did not concern themselves 
with the remote vexation of the Boer President and Council, but 
kept on ransacking the gravel. Early in the year there had 
been some straggling prospecting on the Pniel bank opposite 
Klip-drift, but the first continuous work on a south bank placer 

'"South Africa," George McCall Theal, 1888-1893. "Among the 
Diamonds," 1870-1871. 



THE CAMPS ON THE VAAL 161 

was begun in June by a party from the Klip-drift camp. 1 Their 
undertaking was an unwelcome intrusion on land claimed by the 
Pniel Mission, and the diggers were warned of their trespassing 
by the clergyman in charge. The Mission Station was several 
miles from the diamond placer, and the diggers ignored the 
notice, as they were not interfering apparently with the mission 
work by washing river bank gravel. The placer ground proved 
so rich that the diggers flocked to it rapidly, and the Berlin 
Society which maintained the missions at Pniel and Hebron 
was soon glad to obtain the license fee which it was generally 
able to secure from the diggers on the Pniel field. The pre- 
ferred locations on the Pniel bank were along a stretch in the 
middle of the rising ground, a few yards from the water's edge. 
In this tract diamonds were strewn so continuously as to suggest 
the existence of a flow or stream of them, in the red drift gravel 
between the boulders, to the eye of more than one observer. 
This strip was soon honeycombed with shallow pits reaching 
bedrock about twenty-five feet below the surface. 2 

The flow of prospectors continued to spread until the Pniel 
camp, in a few months, rivalled Klip-drift in size, and the two 
contained a population of four or five thousand people. Small 
stone, brick, and iron buildings for stores and other business 
uses were quickly put up in rows along a main street in the 
heart of Klip-drift camp, which bore the name of Campbell 
Street, and a few others of the same durable materials rose from 
other spots in the fields, but most of the miners continued to 
live in their canvas tents, or in reed huts plastered with clay. 
The stone for building was readily obtained from neighboring 
hillsides, and was neatly cut and laid, so that Campbell Street 
soon compared favorably with any country town street in South 
Africa. Butchers, bakers, and grocers opened shops ; restau- 
rants offered good, plainly cooked food at charges so moderate 
that it was reckoned that a man could be well fed at a cost of 
2s. 6d. a day; a tavern and lodging-house, dignified by the name 
of hotel, accommodated travellers and regular boarders; diamond 
1 "Among the Diamonds," 1870-1871. 2 Ibid. 



i6a THE DIAMOND MINES OF SOUTH AFRICA 



brokers sat ready to judge and buy rough diamonds for export ; 
a music hall had a rude vaudeville show every week-day night; 
members of the Masonic fraternity established a lodge ; and a 
little brick church welcomed all comers to its Sunday services. 1 
Similar buildings were put up less regularly in the Pniel 
camp too, and both sides of 
the river showed the like med- 
ley of iron, brick, stone, light 
wood, and canvas stores and 
dwellings. The first mining- 
town newspaper in South Africa, 




Views of Klip-drift. 

the Diamond News, was started at 
Pniel, a little four-page sheet 
that was chiefly filled with adver- 
tisements of local tradesmen on 
both sides of the river, and the local news and stir of the river 
diggings. Rowboats of an established ferry made regular trips 
across the river from one camp ground to the other, charging a 
passenger sixpence for crossing. So there was easy communi- 
cation, and the two camps were one in their common appearance, 

1 "Among the Diamonds," 1870-1871. " The Diamond Diggings of South 
Africa," Pay ton, 1872. 



THE CAMPS ON THE VAAL 163 

work and sympathy, though the Pniel camp did not pretend to 
the dignity of an independent Republic, but submitted meekly 
to the payment of license fees to the Berlin Mission Society and 
to the assertion of the sovereignty of the Orange Free State, 
represented by a local magistrate, with the adjuncts of a canvas 
jail, whipping-posts, and stocks. 1 

Oddly enough, in view of the shallow gravel bed which was 
the sole support of these camps, the approach of collapse was 
not clearly foreseen. An observer of more than ordinary intel- 
ligence visited the camp at the close of the year 1870, and noted 
the exhaustion of the rich ridge gravel back of Campbell Street, 
where more than two thousand diggers were at work a few 
months before. Yet, while remarking the drift of prospectors 
to outlying placers, he wrote, " Notwithstanding this, Klip-drift 
flourishes, and together with Pniel will no doubt always continue 
to be a head centre of the diamond-digging community." For 
this sanguine view there was some justification in the general 
ignorance of the actual extent of the diamond beds in the alluvial 
deposit, and in the common declaration of a purpose to persist 
in searching for diamonds, even by those whose hard luck 
forced them to abandon the fields for a time. " Hope's blest 
dominion never ends " to the most unfortunate laborer. This 
visitor did not meet one of the many leaving the ground with 
empty pockets who did not protest his resolution to return to 
the diggings in the following March or April after the heat and 
storms of the summer season on the Vaal were past. 2 Fortu- 
nately for these luckless adventurers, there was a new and phe- 
nomenal development of other Diamond Fields, whose output 
soon dwarfed all the returns from the shallow River Diggings. 

1 "Among the Diamonds," 1870-1871. "South Africa," George McCall 
Theal, 1888-1893. 

2 " Among the Diamonds," 1870-1871. 



CHAPTER VI 



THE RUSH TO KIMBERLEY 




HERE was a pretty green valley near the Free 
State settlement of Fauresmith, hardly a mile 
in width, but stretching for several miles to the 
northeast through ridges of volcanic rock kopjes. 
Fauresmith lay in the track of the stream flow- 
ing from the coast ports to the diamond-bearing 
valley of the Vaal, but there was no thought of a probable dia- 
mond field on the plateau so far from a river bed. So for 
months the adventurers passed on without pausing, except for 
a night's camp, on their way to the Vaal. A Boer settler, Cor- 
nelis Johannes Visser, had taken up a considerable part of the 
neighboring valley in his farm of Jagersfontein, where his house 
stood in the midst of a gay, blooming garden. He had died 
before the discovery of diamonds, but his farm was held by his 
widow, Jacoba Magdalena Cecilia Visser, and worked by an over- 
seer in charge. 

A little stream, flowing from the hills, ran through the valley 
in the rainy season, though for the greater part of the year its 
track was only marked by a spruit or dry water-course. De 
Klerk, the overseer, noticed that many small garnets mixed with 
pebbles of agate were sprinkled along the dry bed of this spruit, 
and learned that the diggers on the Vaal believed garnets to be 
an indication of the presence of diamonds. So he began pros- 
pecting one day in August, 1870, on the line of the spruit, awk- 
wardly sifting the dry gravel and sand in a common wire sieve. 
At the depth of six feet he found a fine diamond of fifty carats, 
and the news of his discovery was soon widely spread throughout 
the Free State. 1 

1 "Among the Diamonds," 1870-1871. 
164 



THE RUSH TO KIMBERLEY 165 

His neighbors flocked first to the farm, and the thrifty widow 
Visser was pleased to welcome them, and permitted them to dig 
in her spruit, on allotted patches of twenty feet square, for which 
each paid her a license fee of 1 a month. The phlegmatic 
Boers were not wildly excited by the prospect of fortune hid in 
the spruit, but diamond hunting was an agreeable diversion from 
dull farming, and they came with their wives and children in their 
big canvas-covered wagons, and spread out through the green val- 
ley like country folk at a picnic. The children delighted in their 
search for pretty pebbles and soon filled their pockets with gar- 
nets and agates ; but the digging in the spruit was often so labo- 
rious that the farmers were content to squat on the ground and 
puff their long pipes while their black servants did the digging 
and rock heaving. When natives were not engaged as diggers, 
the farmers and their sons indolently shovelled out the gravel in 
heaps to be sorted by their wives and children. 

Underneath the red surface soil filled with pebbles, there was 
a layer of calcareous clay, varying in thickness from a few feet to 
twelve or more, covering drifts and pockets of gravel thickly 
sprinkled with heavy boulders of greenstone and basalt. It was 
necessary to pry up and tug out these boulders in order to reach 
the underlying gravel, and this task was no child's play. Then the 
gravel was pitched out of the holes, rudely sorted by dry sifting 
in sieves, and picked over by hand in search of the precious 
stones. In some pockets there was quite a sprinkling of diamonds, 
garnets and peridot, mixed with coarse gravel, and the returns far 
exceeded the license charge ; but the diamond deposit was scat- 
tered as irregularly as that of the Vaal River field, and many of the 
workers toiled for weeks on their claims without finding anything 
more precious than the jawbones and teeth of a hyena or jackal. 1 

Attention had hardly been called to the diggings at Jagers- 
fontein when a still more remarkable discovery was made in the 
month of September, 1870, at Dutoitspan, 2 on the farm of Dorst- 

1 "Among the Diamonds," 1870-1871. 

2 The original and correct form of this name was " Du Toit's Pan," or the 
pan or pond of du Toit, the name of the man who first owned the farm. Both Du 
Toit's Pan and Dutoitspan are now used. 



166 THE DIAMOND MINES OF SOUTH AFRICA 

fontein, about twenty miles southeast of Pniel and Klip-drift on 
the Vaal. Du Toit's pan was one of the curious natural land 
basins before described, receiving the wash of the surrounding 
ridges, and holding pools of water during the rainy season and 
sometimes during the year. The title to the farm Dorstfontein 
was granted by the Free State Government to Abraham Pauls du 
Toit on the 4th of April, 1860. Du Toit sold the farm to 
Adriaan J. van Wyk, who had built a little house near the side 
of the " pan," where he was living indifferent to the rush of 
prospectors to the Vaal River, until he was suddenly surprised 
by the finding of diamonds a short distance from his house. 

When the news of this discovery spread, coming, as it did, so 
close upon the revelation at Jagersfontein, there was an instant 
rush of prospectors from the Vaal to the new field, swelled by 
the neighboring farmers and the influx still flowing from the 
coast towns. Van Wyk demanded, at first, a royalty of one- 
fourth of the value of all diamonds found on his farm, from 
every prospector seeking to explore the new field ; but he soon 
concluded to issue licenses at a charge of js. 6d. monthly for 
every allotted claim of thirty feet square. The Orange Free 
State government was aroused to assert its claim of sovereignty 
by the spread of the discoveries, and attempted to restrict the 
allotment of the claims on the farm land, for the benefit of its 
own citizens, by an ordinance prohibiting the issuance of licenses 
to any one except a Free State burgher or farmer ; but this 
requirement was easily evaded at Jagersfontein and Dutoitspan 
by the transfer of licenses granted to Free State citizens. Fur- 
thermore, the spread of the news of the discovery and the result- 
ant rush to the Diamond Fields was soon beyond any possible 
restriction imposed by this little Republic. 1 Van Wyk was pre- 
vailed upon without much difficulty to sell his farm to the 
predecessors of the London and South African Exploration 
Company for ^2600, a fortune far surpassing any glitter of 
pebbles in the ground, in the view of this simple farmer. 

Side by side with the Dorstfontein farm lay the farm of Bult- 

1 "South Africa," Theal, 1888-1893. 



THE RUSH TO KIMBERLEY 167 

fontein, divided by a public roadway. The spread of prospect- 
ing soon passed naturally across the road to Bultfontein and to 
other neighboring farms. Bultfontein was owned by a poor 
Boer, Cornelis Hendrik du Plooy, and before the discovery at 
Dutoitspan a thousand pounds would have been thought a 
grossly extravagant price to pay for the whole farm and its live 
stock. But the luck of van Wyk put a new face on the scrubby 
farm lands near the Vaal, and an eager Free State speculator, 
Thomas Lynch, did not wait over Sunday to buy Bultfontein, 
but amazed the owner by driving out to his farm on the Lord's 
Day, November 14, 1869, an ^ offering ^2000 for his land. Du 
Plooy accepted the offer on the spot, for such a sum in cash was 
vastly bigger in his eyes than any possible return from farming 
or picking up " blink klippe." It is said that diamonds had 
been found on the farm previous to this sale, but Du Plooy was 
not aware of any actual discovery on his land, and preferred cash 
in hand to any gambling chances. The story is told that Bult- 
fontein mine was discovered by the finding of a diamond in the 
mortar used by du Plooy to plaster his house and the subsequent 
search for diamonds in the pit from which the sand had been 
taken. It is true that diamonds were found as reported, but it 
was some time after the mine had been rushed. 1 

On the same day that du Plooy sold his farm to Lynch, 
he was beset by Leopold Lilienfeld and others, who advised 
him that the sale was illegal, being made on a Sunday, and 
eventually Lilienfeld gave du Plooy an indemnity against all 
damages if he would refuse to conclude the sale to Lynch. 
On November 16, 1869, the sale of the farm was concluded 
between du Plooy and Leopold Lilienfeld, Louis Hond and 
Henry Barlow Webb for the sum of ^2000. Hond sold his 
one-third interest to Webb, who, with Lilienfeld, Edgar Eager 
Hurley, and others, formed the " Hopetown Company." 

Lynch brought action against du Plooy for ^10,000 damages, 
and obtained a judgment for ^500 and costs on August 19, 1872. 
In spite of his indemnity du Plooy was then obliged to sue 
i" Among the Diamonds," John Noble, 1870-1871. 



168 THE DIAMOND MINES OF SOUTH AFRICA 

Lilienfeld and his associates, and obtained judgment for ^760 
19^. id. and costs, February 12, 1893. In 1876, when the Land 
Commission heard this case, the London and South African Ex- 
ploration Company had been formed, and the title to the farm 
was granted to that company, as successors of the " Hopetown 
Company." 

Bultfontein was linked to Dorstfontein by the acquisition of 
both farms by one holder, and transfer in a subsequent sale to 
investors associated as the London and South African Explora- 
tion Company. The farm of Vooruitzigt, which lay bordering 
on Dorstfontein and Bultfontein to the north, was bought for 
_6ooo shortly after by other speculative investors, the firm of 
Messrs. Dunell, Ebden & Co., of Port Elizabeth. 

The correct record of these farms is as follows : 

Bultfontein was originally granted by the British Govern- 
ment (then occupying the Free State under the name of " The 
Orange River Sovereignty") to J. F. Otto, December 16, 1848, 
under Warden certificate. 

Dorstfontein was granted by the Free State Government to 
Abraham Pauls du Toit on the 4th of April, 1860. 

Alexandersfontein was granted by the Free State Govern- 
ment to Johannes Cornelis Coezee on the jd of December, 1862. 
That portion cut off by the Free State boundary from Griqua- 
land West was granted to Philip Rudolph Nel and Willem 
Gabriel Nel on the i6th of January, 1880. 

Vooruitzigt was originally a portion of Bultfontein, and was 
sold to D. A. and J. N. de Beer on the i8th of April, 1860. 

At the time of these purchases the price paid for any ground 
outside of a short stretch on the Dorstfontein farm was wholly 
speculative. There had been no considerable discovery of dia- 
monds except along the top of a sloping ridge or long kopje 
lying north, at a distance of about a third of a mile, from du 
Toil's pan. The total area of the three farms was about fifty- 
eight and a half (58*^) square miles. 1 The comparative ease of 

1 The total area of the farms, Dorstfontein (6579 acres), Bultfontein (14,457 
acres), and Vooruitzigt (16,405 acres), is 37,441 acres, equal to 58^ square miles. 



THE RUSH TO KIMBERLEY 169 




Kimberley, 1872. 

working in the new fields was a pleasant surprise to the River 
Diggers, who had been obliged to sink pits in heavy gravel thick 
set with boulders. Now they found diamonds sprinkled through 
a light surface soil of decomposed yellow ground, and many 
stones were so thinly covered with earth that some little brilliant 
crystals were washed free from sand after every heavy rain, and 
lay shining on the ground, to be picked up by sharp-eyed dia- 
mond seekers. 1 The mines were not covered with basalt, but in 
many cases with a layer of rather hard limestone or calcareous 
tufa similar to that which covers a large part of the surface of 
the country in this neighborhood, which has been metamorphosed 
by the evaporation of water charged with carbonate of lime. 

The first swarm of prospectors on the ground supposed that 
the diamonds of Dutoitspan were simply a sprinkling strewn 
through a sand wash like the river-shore deposit. When their 
shovels struck an underlying stratum of limestone with streaks 
of greenish shale, at a depth of two feet or less, they presumed 
that this corresponded to the known bedrock of the placers 
along the Vaal, and had no thought that it was a casing for any 
precious stones. So they simply dug through the soil and 
shovelled the ground into heaps to be sifted dry with common 
wire sieves of coarse and fine mesh. There were no boulders 
in this soil and few large stones, so that their claims could be 
rapidly worked. 2 

The ground contained a plentiful sprinkling of small yellow- 
ish diamonds and some larger stones, but the deposit was so 
shallow that it soon was exhausted. In the course of a week or 
two one digger with the help of a sorter shovelled and sifted all 
the ground of his claim, thirty feet square, and moved to another, 
or rambled off" prospecting over the farm lands. 8 There seemed 

1 '< The Diamond Diggings of South Africa," Payton, 1872. 

2 Ibid. s Ibid. 



1 7 o THE DIAMOND MINES OF SOUTH AFRICA 

no prospect to him that the Dutoitspan ridge still held anything 
to reward the labor of penetrating a rock bed. But after many 
prospectors had ransacked the soil of their claims and abandoned 
them, one of the workers on the ridge or elevated land had the 
fancy to see what might possibly lie under the stratum of lime- 
stone, and determined to cut a few feet, at least, through the rock. 
He found that the limestone soon grew so soft and rotten that 
it could be split easily by the stroke of a pick and the lumps 
crushed by the blow of a shovel. This rotten rock fused soon 
with a curious decomposed breccia of a yellowish color, and 
the sifting of this ground showed, to his amazed eyes, the 
presence of diamonds sparkling on his sieve or on the sorting 
table. 1 

With the spreading of this discovery there came another 
rush of diggers to the ridge that soon covered every patch of 
unoccupied ground on its slopes. Foot after foot the mining 
pits sunk through the soft cement, which was often so decom- 
posed that -the point of a pick pierced it like a mass of dried 
mud. Instead of decreasing in number, the quantity of gems 
in a claim often increased with the deepening of the pits, and 
the proportion of large rough diamonds was far greater below 
the depth of a fathom than in the surface soil or the crust of the 
limestone stratum. Payton says that fragments of volcanic 
rocks green trap and basalt chiefly were scattered through 
the limestone and yellow ground ; but there were very few large 
boulders, and the work of mining was far less laborious than 
any pit-driving in the river bank at Klip-drift and Pniel. 2 

Some cut adits at varying angles in the slope of the ridge, 
and carried out their ground in buckets or wheelbarrows. This 
method of mining shunned the toil of lifting heavy buckets out 
of the pits, but it was dangerous from the frequent ground 
slides and rock falls, and caused many a wrangle when adit lines 
crossed or pits met the tunnels. Others opened their claims 
by cutting a series of descending stages, diminishing in size step 
by step, so that the pit bottom was reached by passing down a 

1 "The Diamond Diggings of South Africa," Payton, 1872. 2 Ibid. 



THE RUSH TO KIMBERLEY 171 

rude rock staircase. This was a rapid and convenient mode of 
opening ground at the start, but where claims were only thirty 
feet square, it was clear that no single claim-holder could go far 
down in this way without reaching a point where the bottom 
step of his staircase would cover the floor of his claim. For this 
reason many preferred to mine more slowly in small perpen- 
dicular shafts, in whose side little niches, familiarly known as toe 
holes, were cut, so that agile men could clamber up and down. 
Or the shaft bottom was reached by means of a knotted rope or 
riem of rawhide, dangling into the pit from a post set in the 
ground near the mouth of the shaft. When a bucket was filled 
with broken rock by a digger working on a pit floor, his mate 
hauled up the load by winding a rope stretching from the handle 
over a rude windlass, or by sheer lifting. When only one 
digger was holding a claim, he was obliged to clamber out of 
his pit and haul up his bucket whenever he filled it. 

To extract the diamonds the broken rock was pulverized 
by beating with shovels and then screened in a- common 
round sieve of coarse mesh, to separate the larger stones 
that were worthless. After this screening the ground passing 
through the coarse wire mesh was carefully sifted, a second time, 
in a rocking sieve of fine, strong wire. This sieve was set in 
an oblong frame, usually about three feet long and two broad, 
with handles at one end and deep notches at the other, gripping 
a narrow strip of rawhide stretched between two upright posts 
called sieve props. When this rocker was swung rapidly, all 
the sand and dust fell through the wire mesh, leaving a concen- 
trate of fine chips and little pebbles of limestone, talc, basalt, and 
trap, carrying a sprinkling of garnets, peridot, and an occasional 
diamond crystal. This concentrate was then taken to a sorting 
table and scraped over in the same way as the river gravel. 1 

Diamond winning on the upland was easier, at first, than 
working the river placers ; but there was one common annoy- 
ance which was much more irritating on the new fields than at 

1 "The Diamond Diggings of South Africa," Payton, 1872. "Among the 
Diamonds," 1870-1871. 



1 72 THE DIAMOND MINES OF SOUTH AFRICA 

the river diggings. Hot winds blew the red dust from the sur- 
rounding veld in clouds over the workers, and these dust blasts 
were mixed with the powdered white limestone and pulverized 
cement of the ridge, shaken through the sieves and blown in 
the faces of the miners, inflaming their eyes, clogging their 
noses, and even coating their skin through their clothes. So 
fine was this powder and so sharply blown that it penetrated 
even hunting-case watches, and few watches could be kept run- 
ning after a month at the diggings of Dutoitspan. 1 

But this was comparatively a trivial concern to ardent dia- 
mond seekers, winning the precious stones so frequently. 
Every day swelled the rush of adventurers to the pan, bargain- 
ing for halves, quarters, and even eighths of a claim on the 
ridge, and roaming over every foot of ground of Dorstfontein 
and the neighboring farms of Bultfontein, Vooruitzigt, and Alex- 
andersfontein in search of new diamond beds. Oddly enough, 
as the prospectors thought, no spot on the whole farm of Dorst- 
fontein rewarded their search outside of the ridge near the pan, 
and for months no better luck attended the hunting for dia- 
monds over the neighboring farms. But where one party of 
the ardent seekers failed to find diamonds, another followed on 
its track and scoured the face of the farms with shovels and 
sieves, with a persistence that was certain to be rewarded, in 
time, if any diamond surface beds existed outside of the ridge 
at Dutoitspan. In the frequent sinking of pits, also, in the 
basins, for water, there was the further chance of piercing some 
hidden bed of diamonds, for the search for springs was hardly 
less keen than the quest for precious stones. 

So, early in i8yi, 2 diamonds were unearthed in the surface 
soil close to the farmhouse of Bultfontein. This discovery was 
followed in the first days of May by the discovery of diamonds 
on de Beer's farm, Vooruitzigt, about two miles from Dutoits- 
pan. 8 Two months later a second diamond bed was uncov- 

1 "The Diamond Diggings of South Africa," Payton, 1872. 2 Ibid. 
3 Ibid. [These dates differ somewhat from those given by Theal and others. 
Payton was on the ground in July, 1871, and his account should be most accurate.] 



THE RUSH TO KIMBERLEY 



173 



ered on the same farm, lying on a gently sloping kopje, at a 
distance from the first location roughly reckoned at a mile. 
This kopje had been searched twice by prospectors, it is said, 
without success, and one report says that the deposit was finally 
discovered through the sinking of a well on the ground. 1 The 
diggers drove their well down seventy-six feet without finding 
water, but at this depth one was amazed to see a diamond of 
eighty-seven carats sparkling on the wall of his dry pit. 

So many conflicting state- 
ments have been made as to 
the discovery of the first dia- 
mond at this location, called 
New Rush or Colesberg Kopje, 
and afterward famous as Kim- 
berley Mine, that I have been 
perplexed to decide to which 
story the most credence should 
be given. The difficulty in ob- 
taining trustworthy data arises 
from the fact that few of the 
original diggers are still alive, 
and that most of those who 
are still living are scattered to 
all parts of the world. More- 
over one cannot always rely 
upon the accuracy of the mem- 
ory of the old diggers now living upon the Fields as to dates 
and details after the lapse of more than thirty years. After 
diligent sifting of all reports and records, however, the following 
conclusion may be said to be well determined. 

Through the courtesy of Mrs. Grimmer, the widow of Dr. 
Grimmer, a practising physician at Colesberg when the Diamond 
Fields were discovered, I was enabled to meet Mrs. Raw- 
storne, the mother of Fleetwood Rawstorne, then (1900) living 
at Cape Town. She is a fine-looking old lady, as her portrait 
1 "The Diamond Diggings of South Africa," Payton, 1872. 




Mrs. Rawstorne. 



174 THE DIAMOND MINES OF SOUTH AFRICA 



shows, well preserved after a long and eventful life in South 
Africa. She was eighty-two years old at the time of our meeting. 
Her memory took her back to the days of the discovery, and 
she related the incidents of the Fields as clearly as if they had 
happened but yesterday. The photograph, here reproduced, of 
the discoverer of Kimberley mine and his party was taken a few 
days after the discovery of diamonds on Colesberg Kopje. 
Fleetwood Rawstorne stands in the middle of the group (page 
175), in the shade of a fine specimen of the camelthorn trees 

which grew upon the mine. They 
had only begun to dig prospect- 
ing holes. The cut on page 176 
shows the primitive method of 
working the diamond - bearing 
ground. I had the story of the 
discovery also from Mr. T. B. 
Kisch, who states that he is the 
only one now living of the first 
four locators. 

Fleetwood Rawstorne, T. B. 
Kisch, and two other diggers 
were prospecting on this kopje 
during the month of July, 1871. 
Some of the party thought they 
saw " indications " of diamond 
deposits, and Rawstorne sent his 
Kafir servant to prospect thor- 
oughly the spot in view. The Kafir returned to his master with 
a diamond of about two carats weight. This discovery was made 
known at once to the other members of the party, and all went 
immediately to the spot and marked and pegged off their claims ; 
Rawstorne pegging three, two as a reward for discovery and one 
as a digger. After the claims had been pegged off Rawstorne 
went to the authorities and reported his discovery. On the 
following day the government surveyor was sent to mark off the 
claims and allot them according to the existing law or custom. 








Mr. T. B. Kisch. (The only one now living 
of the first party who located claims on 
Kimberley Mine.) 



THE RUSH TO KIMBERLEY 




Kimberley Mine just after the Discovery, July, 1871. 

The name of Colesberg Kopje was given to the hillock 
because the lucky diggers, headed by Rawstorne, came to 
the field from the town of Colesberg, near the Orange River. 
The instant flocking of people to the two Vooruitzigt farm 
diggings caused them to be roughly distinguished as " De Beers 
Rush" or " Old De Beers," and " De Beers New Rush," or the 
" Colesberg Kopje " names which endured some months, until 
the " New Rush " was rechristened Kimberley in honor of the 
British secretary for the colonies. 

This inroad of squatting prospectors was greatly vexing at first 
to the owners of the diamond-bearing farms. It disturbed the 
use of the ground for stock-raising purposes, and if there were 
any diamonds on the land, the purchasing speculators wanted to 
hold the beds for their own exclusive development and profit. 
But it was soon evident that this design was impracticable. The 
swarm that covered the ground could not be held in check by 
any force at command of the owners, and stiffly refused to recog- 



176 THE DIAMOND MINES OF SOUTH AFRICA 

nize any assertion of legal claims that took the form of monopoly 
titles. The first diggers on the Bultfontein farm were warned 
off by the owners for trespass. There was a momentary hesita- 
tion till the rush was swelled by numbers so large that the for- 
bidden ground was "jumped" in an hour, and diggers upturned 
the soil to the very door of the farmhouse. Then the owners 
called on the Orange Free State police for help, and the miners 
were driven away for some days ; but the certainty of another irre- 
sistible rush was so ominous that, toward the end of May (1871), 
the proprietors opened the field to all comers on payment of a 
license of ten shillings a month for each claim of thirty feet square. 1 
In the grants of farms by the Dutch East India Company 
there had been no reservation of mineral rights, but from the 
time of the cession to Great Britain, MacNab says the grant of 
lands did not carry a title to " precious stones, gold, and silver," 
which were explicitly excluded, and in 1860 it was enacted in 




Kimberley Mine just after the Discovery, July, 1871. 

1 " The Diamond Diggings of South Africa," Payton, 1872. 
Africa," Theal, 1888-1893. 



< South 



THE RUSH TO KIMBERLEY 



177 



Parliament that no lands containing valuable minerals should 
be considered waste lands of the crown for purposes of sale. 
This did not apply to Griqualand West, as there was no min- 
eral or precious stones act or ordinance in force in this terri- 
tory until Ordinance No. 3 of 1871, of the Orange Free State 
Government. 1 Whether there were mineral reservations in the 
diamond-bearing-farm deeds was 
not questioned by the inrush- 
ing diggers. They would not 
suffer exclusion without a fight, 
but they were willing to pay small 
license charges to the farm own- 
ers for the privilege of working 
allotted claims. The size of these 
claims was fixed by agreement 
with representative " Diggers' 
Committees," chosen by the pros- 
pectors in mass meeting, and these 
committees determined also the 
simple mining regulations and 
camp rules. One committee had 
charge of the Dutoitspan and 
Bultfontein mining camps, and 
another directed the mining at 
De Beers and the Colesberg 
Kopje, pitching its official tent 
midway between these two dia- 
mond beds. 2 Fleetwood Rawslorne. 

The Orange Free State claimed the new diamond fields as 
part of its territory, but its right of control was not vigorously 
asserted in practice. There was a rising issue from the time of 
the discovery at Dutoitspan touching the ownership of the 
district containing the diamond-bearing farms and the diggings 
on the line of the Vaal. The South African Republic claimed 

1 "On Veld and Farm," Frances MacNab, London, 1897. 

2 "The Diamond Diggings of South Africa," Pay ton, 1872. 




178 THE DIAMOND MINES OF SOUTH AFRICA 

the ground north and west of the river, as before noted, but the 
miners at Klip-drift had continued to maintain their rude Republic 
or independent camp, drifting into a condition verging on anar- 
chy, under the doubtful control of a factious " Executive Commit- 
tee," until December 13, 1871, when the camp gladly submitted 
to the authority of a provisional magistrate, appointed by 
Lieutenant General Hay, her Majesty's High Commissioner. 1 
This energetic official had his eyes widely open to the possible 
value and extent of the new diamond-bearing field, and was not 
only disposed to sustain the appeal of the river diggers against 
the monopoly grant of the Transvaal Republic, but wrote 
to President Brand, the head of the Orange Free State, in 
September, 1870, questioning the title of the Free State to the 
Dutoitspan fields and the river diggings at Pniel. 2 

At the time of the creation of the Orange Free State out 
of the domain included in the Orange River Sovereignty, there 
had been explicit recognition of reservations set apart for the 
Basutos, Koranas, and Griquas, native tribes dwelling within 
the limits of the Sovereignty. But there was an apparent lack 
of precision in the reservations or claims of the Koranas and 
Griquas especially, which was accounted of little consequence at 
the time, until the discovery of diamonds, on a tract otherwise 
not worth contesting, aroused rival claimants. The Berlin 
Mission Society claimed the diggings at Pniel on the strength of 
a deed of sale of part of the Korana reserve. Nicholas Waterboer 
and other Griqua chiefs, doubtless prompted by speculative agents, 
set up their claim to a considerable stretch of ground, covering 
Klip-drift and Pniel as well as the upper angle between the 
Orange and the Vaal, containing the diamond fields of Dutoits- 
pan and the surrounding farms. The Orange Free State did 
not dispute the right of the natives to hold such reservations as 
had been assigned to them by the British Government, but con- 
tended that the stretch of the native tribal claims was wholly 
unjustified, and that Pniel and Dutoitspan were clearly within 
the bounds of its domain. 8 

1 " The Diamond Diggings of South Africa," Payton, 1872. 

2 "South Africa," Theal. 3 Ibid. 



THE RUSH TO KIMBERLEY 



179 





Some of the Native Chiefs dealt with by Mr. Richard Southey, Lieutenant Governor of Griqualand 
West, during his Administration. 



i8o THE DIAMOND MINES OF SOUTH AFRICA 



Sir Henry Barlcly succeeded Lieutenant General Hay early 
in 1871 as her Majesty's High Commissioner and Governor of 
Cape Colony, and was expressly instructed by Earl Kimberley, 
the British secretary for the colonies (January 24, 1871), not to 
countenance any annexation of territory outside of the uncon- 
tested limits of Cape Colony, which the Colony would be 
unable to govern and defend with its own unaided resources. 
But the new High Commissioner viewing the situation and 
the course of his predecessor, which he cordially approved - 
replied to his instructions bluntly that the British Government 
" had already gone too far to admit of its ceasing to support the 
cause of either Waterboer or the diggers." 1 He concluded an 
arrangement, accordingly, for the transfer to Great Britain of 
the claims of the native chiefs, subject to the ratification of the 
Home Government, and his representations secured the consent 
of the Ministry, in the following May, to the transfer, and to 
the assertion of British sovereignty over the disputed territory, 
pending the final decision of the special court of arbitration 
which had been convened by the agreement of the contesting 
claimants. 

The court had been opened, in the previous April (1871), 
in the village of Bloemfontein. After considering the evidence 

presented, the 
judges disagreed, 
and the disposi- 
tion of the terri- 
tory depended 
upon the award 
of the referee, 
Lieutenant Gov- 
ernor Keate, of 
Natal. This was 
not rendered un- 
til the 1 7th of October following, and it does not appear that 
the decision was hurried or improperly influenced. But it was 

1 "South Africa," Theal. 




The first Government House and Buildings of the Colony of 
Griqualand West. 



THE RUSH TO KIMBERLEY 



181 



warmly denounced as partial in sweeping aside the claims of the 
Orange Free State and the Transvaal Republic, and confirming 
the alleged title of Waterboer and other native chiefs to a tract 
covering 17,800 square miles, and including the Dutoitspan, 
Bultfontein, De Beers, and Kimberley diamond mines, as well as 




j^. 



\ , I 

s.^i+.^^JfJL**,- 



Sir Richard Southey's Residence, Kimberley. 

the diggings along the Vaal. Four days after this award had 
been made, Sir Henry Barkly proclaimed the grant to the native 
chiefs a part of the British dominions, as the Crown Colony of 
Griqualand West, which was placed under the administration 
of a Lieutenant Governor, Mr. Richard Southey. 1 Thus the 
control of the Diamond Fields was finally determined, and it is 
impossible to doubt that this settlement was greatly contributory 
to the extraordinary advance of diamond mining in these fields, 
as well as to the uplifting and development of the Colonies, and 
to the push of civilization into the heart of the dark continent. 
It has been contended that the award was unjust to both 
of the Boer Republics, and this contention has been supported by 
the citation of a court decision rendered several years later, and 
the allowance of ^90,000 to the Orange Free State by the Lon- 
don Convention of 1876, in compensation for losses sustained 
1 " South Africa," Theal. 



i8i THE DIAMOND MINES OF SOUTH AFRICA 

through the creation of Griqualand West. But it has been 
fairly pointed out by the leading historian of South Africa, 
Theal, an earnest supporter of the rights of the Orange Free 
State and her sister Republic, that the claims of both contest- 
ants were weakly presented at the Bloemfontein court, and that 
Lieutenant Governor Keate cannot be reproached justly for any 
conscious unfairness in deciding the case upon the evidence 
before him, in a manner unsatisfactory to the Republics on the 
line of the Vaal. 

There is, further, the practical view to present of the incor- 
poration of the Diamond Fields in Griqualand West, that this 
was the only feasible solution of the situation, at that time, 
which guaranteed to the irresistible rush of diamond seekers 
from the Cape and all parts of the world a government so 
strong that it could enforce its authority without recourse to 
arms and bloodshed. Klip-drift had already revolted at the first 
preposterous stretch of authority of the South African Repub- 
lic, and maintained its independence until it submitted docilely 
to the British High Commissioner. The seething influx on the 
upland Diamond Fields was clearly on the verge of rebellion 
against any Free State regulations restricting their right of 
entry or supporting any monopoly title. Great Britain, with 
all her array of Imperial power, would not have ventured to 
assert such claims as had been set up by both of the Boer 
Republics, and could not have enforced them without an army 
on the spot. As a matter of fact, she prudently suffered the 
miners to occupy the land without any attempt to maintain 
crown reservations of mineral rights, even after her supremacy 
was undisputed through the formation of the Crown Colony. 
The Boer Republics, on the other hand, would have continued 
to blunder, almost certainly, as they had been doing, if control 
of the Fields had been turned over to them nominally by the 
decision of the referee. 

It did not appear at that time, either, that there was any 
strong desire on the part of the authorities of these Republics 
to assume the cost and responsibility and prospect of collision 



THE RUSH TO KIMBERLEY 183 

which the supported assertion of control of the Diamond Fields 
would have involved. The founders of these states had sought 
only the plain homes of farmers and shepherds on the veld, 
under a government of their own choosing. Neither they nor 
their children were greatly stirred by the uncovering of dia- 
monds, or the prospect of finding more on their lands. They 
disliked the spreading rush to the Diamond Fields, even when 
it was presumed that their own mines were developing. The 
plain, stolid farming folk, stiffly set in their old-fashioned ways, 
had little in common with the sanguine adventurers, delighting 
in stirs and surprises and novelties. Baines tells a story of the 
mobbing of the first surveyor who tried to use a theodolite in 
the streets of Potchefstrom, instead of stepping off the distance 
in the good old way of the " veld-valkt-meester." He avers, 
too, that he was himself made " vogel vrie," " free as a bird for 
anybody to shoot at," for the crime of concealing a sextant 
about his person. 1 This may be a fanciful stretch of fact, but 
there is no doubt of the ingrained conservatism of the Boers. 
How could such a people sympathize with the impetuous and 
ardent spirits that rushed to the Diamond Fields, and what pros- 
pect was there of the docile submission of the one to the other ! 
It can scarcely be questioned, therefore, by a candid observer 
that the conclusion of Lieutenant Governor Keate was the best 
practical settlement, if not the most impartial and accurate. 

It was not to be expected, however, that this significant 
departure from the halting policy of former ministries, this for- 
ward step of Greater Britain into the heart of a region hitherto 
indifferently resigned to the migrating Boers, should be viewed 
with resignation by the embittered Republics whose claims were 
disallowed. Resentment ran so high in the Transvaal that 
President Pretorius was forced to resign. His place was filled 
by a clergyman, Thomas Fra^ois Burgers, and, after the short 
sharp war for independence in 1 880-81, by Stephen J. Paul 
Kruger, a marcher with the Great Trek from the Cape to the 
Limpopo, a lion killer from boyhood as dauntless as David, 
1 " The Gold Regions of Southeastern Africa," Thomas Baines. 



184 THE DIAMOND MINES OF SOUTH AFRICA 

a crafty politician and a religious exhorter, a Covenanter of the 
Covenanters, a Boer of the Boers, uncouth, unschooled, con- 
ceited, bigoted, grasping, bristling with suspicion and prejudice, 
tickled with gross flattery, but a man of iron nerve, intensely 
loyal to his people and their push for independence, self-contained, 
self-reliant, bold, wary, cunning, ambitious, dominating, fore- 
handed masking his plans, biding his time, resolute in action, 
and far-seeing in shaping the future of his Republic. In the in- 
clusion of the precious diamond-bearing province in Griqualand 
West, an inveterate antagonist of British Imperial extension was 
raised to power, whose keen forecast was almost able to over- 
balance the impulse of this great accession to the upbuilding of 
Greater Britain in South Africa. 1 On the coat of arms of the 
Transvaal Republic a lion lay crouching, ready to spring. From 
the day of Krviger's rise to head the Republic, the lion of the 
Transvaal has never shut his eyes nor feared to show his teeth. 

While this protracted controversy for the control of the 
Diamond Fields was dragging on, the rush to the diggings had 
been spreading and moving from the ports of Australia, India, 
and China ; from California, Canada, and the Eastern Atlantic 
states of the American Union from Great Britain and Ireland 
and the countries of Western and Central Europe ; from every 
region of the civilized world, at length, where men of restless 
and sanguine temper were living, who could command the price 
of the passage to diamond-bearing placers, unmeasured in num- 
ber, extent, and richness. The virgin fields of California and 
Australia, once so glittering with gold and so potent in attraction, 
had lost their glamour with the scouring of their sands and the 
passing of their novelties. It had been demonstrated with plain, 
cold figures and dismal accuracy that the average farmer was get- 
ting far more from his wheat or potato patch than the average 
prospector from his scramble in a gold-field. But who could 
calculate, or even pretend to predict with any assurance, the pros- 

1 "South Africa," Theal. "Impressions of South Africa," James Bryce. 
" The Story of South Africa," William Basil Worsfold. " Cecil John Rhodes," 
Biography, " Imperialist." 




STEPHEN J. PAUL KROGER. 
(From a Photograph taken at Kimberley, 1884.) 



186 THE DIAMOND MINES OF SOUTH AFRICA 

pect of fortune in this African wonderland, so phenomenal in 
character and so slightly explored ! Here was a strange, luring 
beacon in the heart of traditional Ophir, where river banks were 
apparently lined with diamonds, where diamonds were strewn 
on the face of farms, where children had diamonds to roll like 
marbles, where wells were driven through diamond beds, and 
huts were plastered with diamond-studded cement. Who would 
not rush to a region so sparkling in promise, so embalmed in 
traditions of resplendent empire, where another Koh-i-nur might 
be lying in wait in the dust for the first passer-by, and where a 
lucky adventurer might stuff his pockets with gems far surpass- 
ing the hoard of any extortionate nabob, and return home with 
a treasure that he could carry as lightly as a full purse ! 

The river placers had not drawn largely outside of the south- 
ern African colonies, but the discoveries at Dutoitspan, Bultfon- 
tein, De Beers, and Kimberley were so unexampled, and the 
mines on the surface were soon shown to be so marvellous, 
that their magnetic attraction was felt all over the globe. Who 
can wonder, then, that the flying, inflated, distorted rumors 
from this African hot-bed puffed up ardent fancy everywhere as 
tongues of flames in tinder, and that men of all nations, call- 
ings, and characters were swept along in the rush to the South 
African Diamond Fields ! Every sailing ship or steamer that was 
bound for a South African port from any part of the world, 
in 1871, bore some adventurers to the new fields. Some 
had good outfits and supplies of money, while others had barely 
been able to scrape together their passage costs. The seamen 
on the ship caught the infectious diamond fever, and ran away 
when the vessels were moored on the African coast, as their 
mates had done, years before, in the ports of California and Aus- 
tralia. Nothing but actual bonds could hold back the diamond 
seekers, and these would not serve if there was any chance to 
cut cords and break irons. 

The swarming of adventurers over mountain terraces, veld, 
and karoo was more motley and ardent than the first rush to the 
Vaal, and every one was consumed by the fear that others ahead 



THE RUSH TO KIMBERLEY 187 

of him were dividing up the rich ground and a day's delay might 
cost him a fortune. So never before was there such a scurrying, 
reckless of lagging ox-teams and horses, blazing suns, and blind- 
ing dust. What a fuming there was, too, on the river banks 
when the sudden floods halted the rush with their impassable 
torrents, and the pilgrims on nettles watched the yellow water 
run surging, swirling, and whirling between them and their goal ! 
Most of the adventurers still plodded along with their bul- 
lock wagons, but some who could afford to pay roundly (.12) 
for transport were carried to the Diamond Fields by the wagons 
of the Inland Transport Company, an enterprising association 




Coach leaving Kimberley for the Coast, 1875. 

which undertook to run a regular coach-line to the Vaal from 
Wellington, the terminus of the short Cape railway in 1870. 
The carriage was a long, narrow wagon with five rows of seats 
for fourteen passengers and a driver. Only forty pounds of 
baggage could be carried by a passenger, but men who were 
anxious to reach the mines were ready to start without even a 
shift of shirts. Eight wiry horses dragged this rattling wagon 
over the rough track at a lively rate, changing teams at relay sta- 
tions, from thirty to forty miles apart, and making the trip to the 
Vaal in eight or nine days when the way was not blocked by 
floods. By this stride of progress the journey from Cape Town 
was made in less than a quarter of the time required by the 
crawling ox-wagons from the other coast ports, although these 



1 88 THE DIAMOND MINES OF SOUTH AFRICA 

towns were two hundred miles nearer the Diamond Fields. This 
was proudly noted as an advance of rapid transit, which prom- 
ised greater developments, and was one of the many stirring 
impulses of the diamond discoveries. But as only one stage- 
coach started weekly from Wellington, the chief contribution of 
the new line to South Africa lay in its promise rather than its 
performance. 1 It was the first push of the enterprise which has 
followed its hoof tracks through the African desert with the tire- 
less race of the iron horse. 

While this swarm was gathering from India, Australia, 
Europe, and America, and pressing toward the diamond mines 
through the southern Colonial ports, another swarm was enter- 
ing the fields from inland Africa. To the native tribesmen the 
opening of the diamond mines was a certain Golconda. For the 
shovelling of gravel under a burning sun, for the heaving of 
boulders, for the shaking of cradles in the midst of whirling dust, 
for the quarrying in pits and the scraping on sorting tables, 
the wiry sinews, pliant muscles, nimble fingers, and sharp eyes of 
Africans, inured to the scorch of the sun, the pelt of the rain, 
and the blast of the sand, were greatly serviceable. So there was 
a cordial greeting of the influx of natives, ready to work for the 
barest pittance of pay while their masters lolled in the shade. 

First came the neighboring Griquas, Koranas, and Batlapins, 
with Basutos from their southern reservation, followed by a 
stream of Zulus, Mahowas, Malakakas, and Hottentots, and 
Kafirs of one hundred tribes, ranging east to the Indian Ocean 
and far northwest into Namaqua and Bechuana lands and north- 
east into Matabeleland and the regions lying beyond the Limpopo 
and the Zambesi. 2 There was every shade of dusky color in this 
throng, from livid and tawny yellow to jet black. Some stalked 
proudly over the veld in the full plumage of the Zulu veteran, 
with flowing ox-tail girdles, armlets, and anklets, decked with 

1 "Among the Diamonds," 1870-1871. 

2 " The Diamond Diggings of South Africa," Payton, 1872. "South Afri- 
can Diamond Fields and Journey to Mines," William James Morton, M.D., New 
York, 1877. 



THE RUSH TO KIMBERLEY 189 

waving feathers and gleaming earrings and bracelets. Others 
vied with this show in greasy red shakos, faded blouses, and other 
cast-off equipments of soldiers and hunters. So the parade ran 
down to the barest loin cloth or utter nakedness, through leopard 
skin wraps, dirty karosses, ragged breeches, tattered shirts, and 
every other meagre covering of the native hunter or shepherd. 
Some of this drift to the mines tramped more than a thousand 
miles over mountain ridges and sun-scorched veld, swimming 
through rivers, scrambling down steep ravines, and plunging deep 
in mud and desert sand, to reach their goal, as many did, gaunt 
skeletons of men, with bleeding feet, and bodies scratched and 
sore and tottering with weariness and hunger. 1 

Diamonds were no temptation to them. They would not 
have walked a mile to pick up a Koh-5-nur. But the white dia- 
mond seekers were willing to pay, for a few months' hunting for 
little white pebbles, enough to buy a cheap gun and a bag of 
powder and balls most precious of all earthly things in the 
eyes of a roving African. Then the white camps were lively, 
humming social resorts, abounding with good food and tempting 
drink, where black men were welcome and well protected. So 
the natives swarmed in faster and faster as the mining progressed 
and the news spread to distant regions. Some of this swarm could 
be persuaded to remain at the mines for a year or more and work 
quite steadily ; but most drifted away, at the end of a few months, 
or as soon as they were able to get their coveted guns and powder 
pouches. Thus while many thousands flocked yearly to the Fields 
from their opening, the outflow kept the supply from swamping 
the demand. As this influx from the dark continent met and 
mingled with the rush from the outside world in the diamond- 
mine workings and camps, how greatly vivid, unique, and stir- 
ring were the kaleidoscopic shifts of this strange concourse ! 
Europe, Asia, Africa, and America had boiled over into a hotch- 
potch, splashed on a diamond bed in the heart of South Africa. 

1 "The Diamond Diggings of South Africa," Pay ton, 1872. "South Afri- 
can Diamond Fields and Journey to Mines," William James Morton, M.D., New 
York, 1877. 



CHAPTER VII 



THE GREAT WHITE CAMPS 



OW quickly and marvellously was the face of 
the little South African stock farms transformed 
by this influx ! Open pasture land, where the 
eye saw one day only a few scattered cattle 
browsing on the thin grass and scratching their 
sides against a stunted camelthorn, was covered 
next day by swarms of roving prospectors, with shovels and 
sieves, upturning grass roots and shaking dry earth through 
their screens. White canvas camps, foaming with life, rose in a 





Kimberley, before the Discovery of Diamonds. 

night, with the seeming magic of Aladdin's palace, at the foot of 
kopjes where, before, a burrowing meerkat was the only tenant. 
Beyond the masses of tents ranged long straggling arches of 
wagon tops and tethered troops of bullocks, horses, and mules. 

190 



THE GREAT WHITE CAMPS 



191 



Only a few months from the day when the first diamond was 
picked up near du Toit's pan, the camp at Dorstfontein was 
proudly claiming the title of the " City of the Pan." A spacious 
market square was laid out on the ground between the pan 




Dutoitspan. (From a very early Photograph.) 

and the ridge covered with diamond diggers, and around this 
square were ranged the white walls of the aspiring camp. 
Streets radiating from the central square gave open access to the 
market-place, and the white tent blocks were soon dotted near 
the square with shops of brick and iron and wood, rivalling the 
pioneer diamond-digging town of Klip-drift on the Vaal. 1 

Klip-drift struggled on with the best face possible, making 
much of its position of vantage as the distributing market of all 
camp supplies from the South African Republic ; but its day of 
ascendancy soon flitted away never to return. In September, 
1871, its chief standard-bearer, the Diamond News, moved to 
the " City of the Pan," and there was no question from that time 
of the preeminence of the " dry diggings," although a rival 
paper, the Diamond Field, bore up for a time under the sinking 
fortunes of Klip-drift. Before the end of the year 1871, 
Dutoitspan boasted " many large hotels," " immense stores," 

1 "The Diamond Diggings of South Africa," Charles Alfred Pay ton, London, 
1872. 



192 THE DIAMOND MINES OF SOUTH AFRICA 




THE GREAT WHITE CAMPS 



'93 



two churches, a hospital, and a theatre, and might have men- 
tioned, besides, its less distinguished billiard room, "canteens," 
and dance halls. 1 It was surely a wonderful birth of a smartly 
growing infant city on the face of scrub-covered prairie in the 
heart of South Africa. 

The rise of the camps at De Beers and Kimberley was even 
more rapid than the growth of the camp on Dorstfontein and 
Bultfontein farms. There was no regular working in the De 
Beers diggings before May, 1871, but the diggers could buy 




Kimberley, 1871. 

Christmas presents that year in rows of brick and iron stores 
on the main roadsides, intermingled with " hotels " and saloons, 
and a great white canvas town was spread out in a picturesque 
medley of tents and marquees, straggling far over the veld, and 
seeking the shelter of some stubbornly rooted mimosa or camel- 
thorn. 2 Kimberley's growth was still more surprising. Three 
months after the rush began, the Colesberg Kopje was the centre 
of an immense encampment in whose heart streets were irregularly 
laid out, and neat stores built of iron and brick. In December, 
1871, there were, by actual count, on the lower street of Kim- 

1 "The Diamond Diggings of South Africa," Payton, 1872. - Hid. 

o 



194 THE DIAMOND MINES OF SOUTH AFRICA 

berley, six stores, four hotels, and several butcher and shoe- 
maker shops, besides a billiard room and saloon. On the upper 




Kimberley Mine, 1871. 

or main street there were three hotels, several diamond merchants' 
offices, a wholesale spirit and provision store, a bakery and con- 
fectioner's shop, a drug dispensary, butchers' shops, eating houses, 
bars, club and billiard rooms, and other miscellaneous shops and 
resorts. On the edge of these white-walled cities, and on the 
slopes of all the neighboring hills, were scattered the huts of wood 
or dirty canvas or mud-plastered stones, where the native blacks 
huddled together. When even this cover was lacking, some 
slept in tents, or in burrows scraped in the hillsides. How many 
diamond seekers were massed in these camps at the height of 
the rush can hardly be reckoned with any approach to exactness. 
There may have been fifty thousand whites and blacks on the 
Fields, for the flow to Dutoitspan is said to have mounted as high 
as forty thousand shortly after the opening of the Vooruitzigt 
farm mines. 



THE GREAT WHITE CAMPS 195 

When, after long weeks of plodding over rugged mountain 
ranges, parched karoos, and rolling prairie, a traveller suddenly 
saw rising before him these white camps, springing up like pro- 
digious mushrooms in an African desert, even the dullest brain 
was strangely disturbed. It was hard to realize that these exotic 
plants were the work of men's hands, for they seemed rather the 
fantastic conceit of the trance of an opium eater. Here were 
such cities as the mirage shapes from clouds or as Solomon 
might have built with the help of his docile genii. When they 
lay outstretched and gleaming under the burning sun in the 
full splendor of noon, they were weird creations to amaze the 
beholder ; but who can conceive their impress at night, under 
the towering sky dome sprinkled with stars, with their masses of 




Around Kimberley Mine, 1871. 

twinkling and sparkling lights on the black face of the veld, like 
the tail of a fallen comet. 1 

1 "The Diamond Diggings of South Africa," Payton, 1872. "To the Cape 
for Diamonds," Frederick Boyle, 1873. "South African Diamond Fields," Will- 
iam Jacob Morton, 1877. 



196 THE DIAMOND MINES OF SOUTH AFRICA 

Outside of these three main camps tents were thickly 
sprinkled around the farmhouse of Bultfontein, in a field where 
a thousand diggers were at work in the first week of the rush, 
after the ground was opened in May, 1871. Immediately south 
of this diamond-bearing farm was the farm Alexandersfontein, 
where many prospectors were also turning and sifting the ground. 
By the determination of the limits of Griqualand West these 
diggings, as well as the chief camps, became part of the British 
Colonial domain ; for the boundary line separating the new Col- 
ony from the Orange Free State ran just outside of this cluster 
of farms, Vooruitzigt, Dorstfontein, Bultfontein, and Alexan- 
dersfontein, through the outlying farm of Benaauwdheids- 
fontein, where no diamond mine had, as yet, been discovered. 1 
So all the known diamond fields of South Africa, except the 
Jagersfontein farm within the bounds of the Orange Free State 
and the shallow Vaal River placers, were bunched on a plateau 
four thousand feet above the sea level, within the angle formed 
by the junction of the Vaal with the Orange River, on a patch 
with a radius of 1.72 miles at the crossing of longitude 24 46' 
east of Greenwich with latitude 28 43' south of the equator. 

The London and South African Exploration Company, by 
its purchase of Dorstfontein, Bultfontein, and Alexandersfontein, 
held a tight grip on the mineral rights comprehending the dia- 
monds on all these farms, and leased the surface diggings under 
licenses of IQJ. for every claim 30 feet square. Messrs. Dunell, 
Ebden & Co., of Port Elizabeth, held the farm of Vooruitzigt, 
and exacted the same license fee for working claims which were 
laid out in squares 30 by 30 Dutch feet, or 3 i by 3 1 English 
feet. 2 Outside of the Colesberg Kopje or Kimberley mine all 
the diggings were at first a jumble of holes, pits, and burrows, 
with no attempt to secure any system or union in mining. But 
the objections to this helter-skelter opening of the ground were 
so apparent that a strict reservation of roadways to give access 
to all parts of the surface of the mine was insisted upon by the 

1 "Diamonds and Gold in South Africa," Theodore Reunert, 1893. 

2 "The Diamond Diggings of South Africa," Payton, 1872. 



THE GREAT WHITE CAMPS 



197 



Orange Free State Inspector of Mines, in the laying out of 
claims on the Colesberg Kopje. His authority was then so far 
recognized that his direction controlled the survey and opening, 
on July 21, 1871, of the diggings since famous as the Kimberley 
Diamond Mine. 

Roadways, 15 feet in width, running approximately north 
and south, were carried across the longer axis of the diamond 
bed, at a distance of 47 feet from one to the other. Kach road 
cut 7.', feet of surface ground from the side of the bordering 




Kimberley Mine, 1872. 

claims, so that the working surface of each allotted claim was 
31 by 23^ feet. Fourteen of these roadways crossed the mine, 
whose ground surface permitted the laying out of about 430 
claims of the allotted size, 3 1 feet square. A great many more 
claims had been granted to license-holders before the survey, 
for there had been no accurate measurement of the kopje, and 
there was a consequent overlapping and conflict of locations and 
spreading of claims beyond the limits of the diamond-bearing 
ground. In the settlement of contests the claims were split up 
by concessions, bargains, and sales, until there were not less than 
1600 separate holdings of claims, and fractional parts running as 
small as -3^, or about 7 square yards. A lucky claim-holder 



198 THE DIAMOND MINES OF SOUTH AFRICA 

would sell off parts of his claim or the whole at high prices; for 
bidders were ready to pay large premiums beyond the license 
fee of los. exacted from every working owner, whether his claim 
was full size or a paring. The competition for a share in the 
riches of the ground was only less keen at De Beers, and there 
was a like subdivision of claims there, and not infrequently at 
Dutoitspan and Bultfontein. 1 

It was obvious from the start, without any stretch of fore- 
sight, that these minute subdivisions of claims and individual 
working were only practicable in open cuttings whose depth 
must depend on the character of the ground and the cooperation 
of the miners. But at the outset of the mining in these Fields 
no one could forecast the unknown continuance in depth of the 
diamond deposits, and few supposed that the new beds differed 
essentially from any before uncovered, and were vastly more im- 
portant than the shallow gravel wash along the banks of the 
Vaal. It was commonly expected that some barren stratum 
would be reached not far from the surface, corresponding to the 
" bed rock " of the river diggings, and that this must terminate 
the hope of the diamond seekers. 2 So the rush for the surface 
claims was the keener, in view of the belief that a few months' 
work at most would exhaust the precious deposit, and nobody 
paused to consider what he would do if he was unable to sink 
an open pit deeper. 

Beneath the red surface soil at Dutoitspan a thin layer of 
calcareous tufa 3 had been exposed, below which lay the dia- 
mond-bearing breccia which the miners called " yellow ground " 
from its prevailing color. 4 At De Beers and Kimberley there 
was comparatively little limestone beneath the red soil, for the 

1 "The Diamond Diggings of South Africa," Payton, 1872. 2 Ibid. 

3 I look upon the calcareous tufa which covered the diamond mines as only 
the altered yellow ground which had been metamorphosed by the evaporation of 
water highly charged with carbonate of lime. The calcareous tufa which covered 
the Premier mine was diamond bearing. This is the only one of the mines 
whose surface ground has come under my personal observation. 

4 "The Diamond Diggings of South Africa," Payton, 1872. 



THE GREAT WHITE CAMPS 199 

rich " yellow ground " rose nearly to the surface under a thin 
coating of chalk. It appeared in exploring the yellow ground in 
most of the openings that the deposit was enclosed in an oval- 
shaped funnel of shale, or decomposed basalt resting on shale, 
which the miners called " reef." This reef contained no dia- 
monds and marked the limits of any profitable prospecting. 
The surface area of the yellow ground within one of these fun- 
nels ranged from about ten acres at Kimberley to twenty-three 
acres at Dutoitspan, and on these patches all the diamond-bearing 
claims of the Fields were located. 1 

When the bottom of the " yellow ground " was reached at a 
depth of from fifty to sixty feet below the surface, it was sup- 
posed at first that diamond digging in the funnels had come to 
an end ; but the hard underlying rock was cut by experi- 
menters, and it was found, to the delight of the miners, that 
this also was diamond bearing. It was a breccia composite, 
essentially like the " yellow ground " above, but much more 
compact and hard, and of a prevailing bluish slate color, so 
that it was familiarly known as " blue ground." Exposure to 
the air, sun, and rain decomposed it so rapidly that most of the 
rock could be readily pulverized after a few weeks, and its 
precious contents extracted by sifting. The whole mass of the 
ground in the funnels was diamond-bearing, in greater or less 
extent, except in occasional streaks and masses of barren shale, 
floating reef, floating shale, or non-diamond-bearing volcanic 
mud, and volcanic rocks. So the pit sinking was widened to 
the extreme limits of the claims, and the entire area of yellow 
and blue ground excavated in open quarries. 

The work was pushed with feverish energy and remarkable 
rapidity in view of the bare hand labor and crude mining appli- 
ances, but there was no uniformity of method or extended 
cooperation. Every claim-holder cut down his patch with pick 
and shovel, and lifted the broken ground in a way that suited 
his individual notion. Some set stout windlasses in the surface 
ground near the edge of their claims, and hoisted buckets filled 

1 "The Diamond Diggings of South Africa," Payton, 1872. 2 Ibid. 



200 THE DIAMOND MINES OF SOUTH AFRICA 




THE GREAT WHITE CAMPS 201 

on the working levels. Others carried up buckets and tubs 
and rawhide sacks to the surface, climbing ladders resting on 
successive terraces, or mounting flights of steps cut in the rock, 
or trundling wheelbarrows up plank inclines. Around the edge 
of the mines there was a mustering of carts, and barrows, and 
carriers, to bear off the quarried ground to depositing places, 
where it was dried, pounded, and sifted. 1 

The open quarries, swarming with workers, buzzed like pro- 
digious beehives. The upsetting of the tower of Babel would 
scarcely have poured out such a medley of tongues and sounds. 
From the vast amphitheatres scooped in the rock there rose in 
the air the clicking of picks, the rasp and clatter of shovels, the 
cracking of rock, the rattle of gravel, the thud of bucket-filling, 
the creaking of windlasses, the tramp over planks, the thump of 
wheelbarrows, the rolling of carts, the lowing of bullocks and 
braying of mules, mingled with calls and chatter and chants 
of whites and blacks in an indescribable din. Diggers in rough 
working dress, and natives almost stark naked, bent and heaved, 
and scrambled and climbed, side by side, reeking with sweat 
and grime, in an ever shifting, restless swarm that covered the 
face of the quarry like flies in some monstrous sugar bowl. The 
flocking in of the native African tribes joined with the white 
diamond seekers in opening the strange funnels of crystal- 
sprinkled breccia made a compound of color, feature, and 
character never before assembled in any mines on the face of 
the earth. 2 The sinewy negroes proved themselves such willing 
and sturdy workers in the dust and heat of the sun-scorched 
quarries, that the claim-holders were glad to hire them and 
confine their own work to the task of overseers, directing the 
digging and hauling, and the sifting and sorting. No blaze of 
the sun and no whirl of the dust could subdue their bubbling 
spirits, breaking out in wild whoops and chants, and yelling in 
pack when any big diamond was found, revelling in every 
chance diversion, the fall of a bucket, the slip of a ladder, the 

1 "The Diamond Diggings of South Africa," Payton, 1872. "South Afri- 
can Diamond Fields," Morton, 1877. 2 Ibid. 



202 THE DIAMOND MINES OF SOUTH AFRICA 




THE GREAT WHITE CAMPS 203 

tumble of a climber, and convulsed with laughter whenever they 
could set mules capering, or bullocks shying or balking by shrill 
whistles and screams, and mimicry of a driver's call: "Yek!" 
and " Trek ! " " Ah now ! " and " Whoa ! " and so through the 
range of cries, Dutch, English, and African. 1 

Almost all the natives were barefooted, and most were bare- 
headed, barebacked, and barelegged, except in the coldest weather. 
Some had ragged trousers, and others ragged shirts, but few put 
both on together. A greasy, gaudy handkerchief twisted around 




Kimberley Mine, 1872. 

a black head, and party-colored bunches of rags, or moochies 
made of the tails or skins of wild animals, dangling from a waist- 
belt of rawhide, were a camp parade dress too precious to use 
in the quarries. Mingled with these wild Africans, the white 
miners worked soberly and arduously, bearing the pains of 
diamond digging stoically, in the hope of its rewards. Their 
working clothes were commonly plain suits of brown corduroy 
or other coarse cloth adapted to the season, and when the sun 
shone they wore generally broad-brimmed straw hats, or pith 
helmets, with light muslin " puggarees." 2 

It was long before there was any notable advance in the pro- 

1 "The Diamond Diggings of South Africa," Payton, 1872. 2 Hid. 



204 THE DIAMOND MINES OF SOUTH AFRICA 

cess of separation of the diamonds from the ground, beyond the 
cradle for dry sifting, constructed to take the place of the com- 
mon hand-sieve at Dutoitspan. Level spots were sought on 
the veld near the mines, or patches of ground were levelled 
sufficiently to serve as dumping places, where the broken dia- 
mond-bearing breccia was piled and spread out. The "blue 
ground " exposed to the air crumbled away by degrees, but the 
miners were rarely patient enough to wait for this disintegration, 
preferring quick returns by pulverizing the ground with their 
shovels and mallets. This was hard work and costly, from the 
loss in imperfect pulverization. But the diamond seekers were 




Roads in Kimberley Mine, 1871-1872. 

poor men who could scarcely afford to hold any stock of blue 
ground for the sake of increased returns, even if they had been 
able to guard their depositing floors from theft. After pound- 
ing the broken rock it was sifted in the midst of dust clouds by 
rockers swung on riems of rawhide, and the concentrate was 
then scraped over and sorted. 1 In July, 1871, a large cylindri- 
cal revolving sieve, driven by a small steam engine, was put at 
work by some American miners, and this sifting machine was 
said to be an efficient and rapid separator. The pulverized 
ground was thrown into the upper end of the screen, which was 

1 "South African Diamond Fields," Morton, 1877. "Diamonds and Gold 
in South Africa," Reunert, 1893. 



THE GREAT WHITE CAMPS 



205 



rapidly revolved, and the concentrate passed out through the 
lower end, falling upon a sorting table. The cylinder, covered 
with fine wire mesh, sifted out the dust thoroughly, and its opera- 
tion was so rapid that thirty cartloads of diamond-bearing ground 
were screened daily. Its owners claimed to be able to sift all 
the ground in a claim thirty feet square to a uniform depth of 
thirty feet in three weeks. The machine attracted a curious 
crowd at first, when the steam whistle blew off and the cylinder 
began to throw off thick clouds of dust, but for some reason its 




Kimberley Mine. (Showing workings in 1872. Subsidence cracks appear in the foreground.) 

use was not long continued. Probably the fine mesh was too 
light to bear the strain and friction of the revolving rock 
fragments. 1 

The amount of ground which any one man could work, was, 
of course, very small, but there were so many workers on the 
Fields that the aggregate extent of ground sifted was enormous, 
and the breccia in spots was so thickly sprinkled with crystals 
that many miners won rich rewards. When Payton was leav- 
ing the field in November, 1871, it was estimated that from 
forty to fifty thousand pounds' worth of diamonds were taken 
1 " The Diamond Diggings of South Africa," Payton, 1872. 



206 THE DIAMOND MINES OF SOUTH AFRICA 

weekly from the Colesberg Kopje alone, and he states that the 
best claims had risen in value from ^100 or less to ^4000^ 

It was clearly shown, too, that even the highest price paid 
for a claim might be cheap, for one poor Dutchman, " Smuts," 
who bought half a claim for ^50, is said to have found dia- 
monds in two months' working to the value of ^1 5,000 or 
more. Another digger found, in a few months, no less than 
730 stones in his claim, one of which weighed 156 carats. 2 
Such great good fortune was rare in the other mines, and many 
miners won little or nothing from months of hard work in their 
claims, but in the Colesberg Kopje, or Kimberley mine, the 
prizes were so common and exciting that every foot of ground 
was covered by diamond seekers. When the rubbing of shoul- 
ders was too close for comfort, one or more of the partners in 
a claim would be pressed to sell out and start again prospect- 
ing. Sometimes a share in a claim, worth many hundreds of 
pounds, would be risked on the toss of a penny. 8 

In the heat of the search and extraction many fine diamonds 
were fractured, and many of the smaller stones ran through the 
sieves into the tailings, as was afterward demonstrated when 
the waste heaps were reworked with better appliances. 4 The 
Kimberley mine produced some stones of large size, running 
sometimes over one hundred carats, but the mass of crystals ran 
under five carats. A yellowish tinge was more marked in the 
diamonds of the uplands than in the river stones, and many 
otherwise superb crystals were so decidedly " off color " that 
their value was greatly impaired. 

It was early noticed that the diamonds of one mine often 
differed materially from those of another, and even in the same 
mine diamonds of one section were unlike the yield of another. 
Thus, in the west end of the Kimberley mine the diamond crys- 
tals were exceptionally perfect octahedrons, or exceptionally white 
" glassy stones," as the miners called them ; while elsewhere in 
the mine the crystals had,more commonly, rounded and bevelled 

1 " The Diamond Diggings of South Africa," Payton, 1872. * Ibid. * Ibid. 
4 "Diamonds and Gold in South Africa," Reunert, 1893. 



THE GREAT WHITE CAMPS 



207 




ao8 THE DIAMOND MINES OF SOUTH AFRICA 




Kimberley Mine, 

edges, and, more or less, a yellowish tinge, and there was a large 
proportion of split, flawed, and spotted stones, and boart. The 
De Beers mine crystals resembled the Kimberley stones, but 
their quality ran a little below the Kimberley mine. Dutoitspan 
produced comparatively few stones, but the average weight was 
notably large, and the crystals were of fine color. Bultfontein 
stones differed greatly from those of the other mines. Here 
the diamonds were chiefly small, rounded octahedrons, many of 
them so pocked and spotted that the crystals had a cloudy 
appearance. 1 These crystals were greatly inferior to the " glassy 
stones " of Kimberley or the large diamonds of Dutoitspan ; 
but the Bultfontein surface ground yield was so uniform at 
first, that many diggers held and worked claims for the sake of 
sure, if small, returns to defray their expenses, while they counted 
on their Dutoitspan claims for the occasional large stones that 
richly rewarded a lucky digger. 

All the crystals in the blue ground were encased in a smooth 
bed of the same material which did not adhere to the diamonds, 
so that their lustre, when extracted, was quite bright or glassy. 

1 "South African Diamond Fields," Morton, 1877. 



THE GREAT WHITE CAMPS 



209 







1872, looking South. 

Amid the mass of white and light yellowish stones in all the 
mines were scattered some of varied color. Brown was the 
most common of these ; next came the deeper yellow shades, 
and pale blue stones were sometimes uncovered, as well as the 
black diamond (boart) used for setting drill-crowns. Pink, 
mauve, and green diamonds were occasionally found, but were 
less common than in the river diggings. 

As already mentioned, it has been estimated that the rush 
which built up these mining camps and covered the surround- 
ing farms with prospectors brought fifty thousand men to the 
new Diamond Fields in the first year, though the shifting popu- 
lation of the Fields did not rise as high as that at any one time. 1 
The influx of native Africans was not so large at first, but 
increased from year to year. Morton says that there was a flow 
of thirty thousand natives annually to the field for seven years 
after the discovery of the mines. 2 This is a credible estimate, at 
least, in view of the constant drifting away from the field of the 
native workers, after a few months' stay, when they had earned 

1 "The Diamond Diggings of South Africa," Payton, 1872. 

2 " South African Diamond Fields," Morton, 1877. 
p 



210 THE DIAMOND MINES OF SOUTH AFRICA 

sufficient money to buy their coveted guns and ammunition, and 
wives, cattle, blankets, etc. 

The bulk of the general merchandise was hauled slowly from 
the coast ports in ox-wagons. Algoa Bay (Port Klizabeth) was 
the chief port of supply at first, and the transport to the Dia- 
mond Fields was a trip ranging from thirty days, at least, to six 
months. 1 Certain kinds of food beef, mutton, poultry, game, 
dried venison, commonly called " biltong," 2 and maize meal 
(mealie meal) were furnished quite cheaply and plentifully from 
the neighboring Free State and the South African Republic, 
through the Klip-drift distributing market. Tobacco, butter, 
eggs, and honey were less freely supplied from the country, and 
commanded a ready sale. Ordinary beef and mutton sold for 
4</. a pound in 1871, with an additional charge for choice steaks. 
A whole sheep could be bought at wholesale for 4^. Game, 
chiefly springbok, blesbok, and wildebeest, was as cheap as 
mutton. Chickens and ducks ranged from 2s. 6d. to ys. 6d. 
apiece. The price of eggs ran high, ranging from is. 6d. to 4^. 
a dozen, and butter was sold at from is, 6d. to 5^. per pound. 
For " Boer meal," a coarse wheat flour, the charge was from 
35-T. to $os. per muid, about 183 pounds ; white flour brought 6d. 
a pound; rice 9^.; sugar and tobacco 9^. to is. ; oranges and 
onions were sold at IQS. per hundred, and dried fruits at from 
3*/. to yd. per pound. 8 

The most urgent calls were for fresh vegetables, and the 
supply was so meagre that the prices shot up to exorbitant fig- 
ures. From 5-f. to js. was freely given for a bucketful of 
potatoes, and the wholesale price for a bag of a hundredweight 
was from ^2 to 1 ids. Haifa crown (sixty cents) was often 
paid for a small cabbage or a handful of onions. Choice forage 
for the horses and mules was almost as costly as vegetables. A 
bundle of five pounds of unthreshed oat hay was sold for as high 

1 "The Diamond Diggings of South Africa," Payton, 1872. 

2 Biltong is made of meat of any of the antelope species, but that made from 
the springbok is considered the best. 

3 "The Diamond Diggings of South Africa," Payton, 1872. 



THE GREAT WHITE CAMPS 211 

as 2J. Dry cut fuel was as high priced as forage. Bundles of 
light sticks sold from yd. to is. per bundle, and ^3 was charged 
for a load of good firewood. There was a considerable forest 
growth on the hills near the Vaal River, and many thickets 
on the ridges nearer the camps, but the cost of cutting and haul- 
ing was so great that many diggers contrived to make their fires 
of dried bullocks' dung (buffalo chips as they were called by the 
emigrants crossing the American plains), as they had learned to 
do when crossing the karoo. 1 

Market auctions were the common and popular mode of 
selling food and ordinary miners' supplies. Criers swinging bells 
rang up the drowsy camps for the early morning market, where 
meat, eggs, butter, fruit, and vegetables were offered from wagons 
and stalls in the open market squares. These sales and gather- 
ings of bidders and lookers-on formed one of the liveliest camp 
scenes, especially on Saturday, when thousands of whites and 
blacks flocked to the auctions, surrounding the stands with dense 
masses of jovial bargainers. How strange and curious to a 
newcomer's eye was the market show, carcasses of big brown 
shaggy wildebeests hanging up in line with sides of beef, ante- 
lopes with slender legs stretched out stifHy among the slaughtered 
sheep and lambs, strips of biltong and freshly killed kids, 
little long-legged hares, party-colored bustards, red-wing par- 
tridges, red-legged plovers, guinea fowl, ducks, geese, and other 
wild fowl, mingled with the poultry from country farmyards ! 
Here were lines of huge tent-covered wagons filled with hides, 
and wool, and meal, and wood, driven to market by the stolid 
Boers or Hottentot servants grinning from ear to ear. Potatoes, 
and beets, and carrots, and onions, and cabbages were piled in 
heaps, tempting the last shilling of scurvy-haunted men. The 
gobbling of turkeys, the crowing of cocks, the quacking of 
ducks, swelled the chorus of chatter and laughing and singing 
and badinage, that smothered, at times, the brisk calls of the 
auctioneers and the offers of the diggers and the hotel and shop 
keepers. 2 

1 " The Diamond Diggings of South Africa," Payton, 1872. 2 Ibid. 



212 THE DIAMOND MINES OF SOUTH AFRICA 




THE GREAT WHITE CAMPS 



213 



In the afternoons special sales of tents, miners' tools, guns, 
and general merchandise were frequently made by auction, and 
large stocks were sometimes sold off completely in this way. 
Often in the flurry of competition these goods brought absurdly 
high prices, when the market was overstocked with like articles 
in the stores. It was observed as a curious fact that scarcely a 
bid could be got for revolvers, which many adventurers had sup- 
posed to be an indispensable part of their outfit. There were 
very few outbreaks of ruffianism in the camps, where the great 




Markel 



body of miners was disposed to be orderly, and occasional sprees 
were the chief disturbances. The swaggering bullies, and cheating 
gamblers, and lurking garroters, who infested the seething camps 
of Nevada and Colorado, rarely drifted as far as these isolated 
Diamond Fields, and the few who came in were held in check. 

The crying need of the camps was good water. The Du- 
toitspan basin did not always hold out through the dry season, 
and besides, this pan was filled by drainage and was not whole- 
some ; but two rude dams were built that served to store up 
drainage water longer than the natural reservoirs. To eke out 
the supply the " Diggers' Committee " at Dutoitspan and Bult- 
fontein sunk several wells which furnished some additional water, 



214 THE DIAMOND MINES OF SOUTH AFRICA 

and a digger was licensed to draw two bucketfuls daily upon the 
payment of one shilling a month for his water rights. This 
privilege was so keenly sought for that there was always a little 
crowd of men with buckets, waiting their turn, at the mouth of 
a well in the daytime. The water was muddy, but it was never- 
theless eagerly drunk, and the stinted supply was too precious for 
washing. Following this push of the committee, prospecting 
water shafts were sunk by private enterprise, and when water was 
reached, the well was opened to a limited number of subscribers 
upon payment of a monthly fee of four shillings. 1 

At Kimberley, water, for months, was so dear that it was sold 
for threepence a bucket, and a daily washing of face and hands 
was a stretch of luxury. A stinted bath at Dutoitspan cost two 
shillings and sixpence, and bathing at the other camps was 
rarely attempted. When the coating of grime grew unbearable, 
the best resource was a ride or tramp to the Vaal and a plunge 
in the river. In the dry season, when the air was full of floating 
dust from the claims and cradles, and when hot winds from the 
veld blew in clouds of red sand, the dearth of water was bit- 
terly felt, and no joker was safe who ventured to recall the " old 
oaken bucket " and other vain visions of cool, bubbling springs. 

Often the dust-storms passed beyond the aggravation of 
thirst and discomfort, driving sand-whirls so furiously in the 
faces of the workers that the hardiest men were forced to drop 
their picks and shovels, and buckets and cradles, and run to 
cover. Then, for hours, storms would rack the tents, straining 
every cord and stitch of canvas to the snapping point, and often 
tearing rents in the walls, or pitching over tent-poles and all in 
utter wreck. Even when the stout posts, braced and guyed 
against a hurricane, bore the strain unyielding, the sheltered 
miners had to swelter in a mist of dust that was blown through 
the crevices into every fold of bedding and clothing, and coated 
every inch of their skins with irritating powder. 

Next to this pest of dust was the plague of flies, little and 
large, black and green, that swarmed over the camps in countless 

1 "The Diamond Diggings of South Africa," Payton, 1872. 



THE GREAT WHITE CAMPS 215 

myriads in the summer season, tainting every morsel of food, and 
settling on every bare face or body with a dash so bold and per- 
sistent, and a grip so malignant, that it hurt like a sting. No 
possible device could clear the tents completely, or keep out 
these swarms ; but the miners armed themselves with big whisks 
of wildebeest and ox tails, and got some relief by constantly flick- 
ing and slashing, or when they were forced to use both hands 
at work on the cradles or sorting tables, "fly flappers" stood by 
to brush back attacks. 

Hot days in the dry diggings on the bare veld were more 
keenly felt than the same days on the tree-fringed Vaal, and some 
midsummer days were too scorching even for the endurance of 
the seasoned black diggers. But, except at midday, few work- 
ing hours were lost when the sun was shining. The swooping 
thunder-storms were scarcely less terrific than the storms in the 
river valley, striking the camps with drenching pelts of rain and 
heavy hail, hurled from cloud banks blazing and bellowing with 
monstrous forks of lightning and stunning thunder peals. 

The clear winter days were greatly invigorating. At break 
of day it was often so cold that jugs of water were skimmed 
with ice and a hoar frost covered the ground. But when the 
bright sun mounted the sky, the chill air was so warmed in a 
few hours, and so pure on the breezy veld, that the miners 
gained fresh spirit with every breath, and went through their 
monotonous round of work with unflagging life and good humor. 
The actual record of a week at the mines, in August, 1871, gives 
a clear idea of the winter shifts of temperature. 1 

DAY. NIGHT. 

Highest Temperature. Lowest Temperature. 

Aug. 21 83 Fah 40 Fah. 

" 22 85 " 35 " 

" 23 83 " 30 " 

" 24 92 " -. 33 " 

" 25 93 , 28 " 

" 26 56 " . . . . . \ 28 " 

1 "The Diamond Diggings of South Africa," Payton, 1872. 



216 THE DIAMOND MINES OF SOUTH AFRICA 

The health of the camps was usually good, except in mid- 
summer, when " low " fever, diarrhoea, dysentery, and colic were 
prevalent. The impure drinking water was the most persistent 
cause of sickness and the most difficult to combat. Inflamma- 
tion of the lungs from the fretting dust, and mild scurvy, were 
the other common ailments, occasioned by the conditions of life 
at the diggings. 1 

It was not all work and no play in these diamond diggings. 
Saturday afternoon was commonly taken as a half holiday in 
addition to the Sunday rest and recreation. In the springtime, 
or the beginning of the rainy season, fresh flowers sprang into 
bloom on the face of the veld, and birds built their nests in the 
grass and thickets. Little dusky black-and-white birds, recall- 
ing the English linnet, were sweetly trilling songsters, and were 
so fearless and sociable that they flew everywhere over the dig- 
gings, in the midst of the dust and stir, perching on heaps of 
broken rock, or even on the diggers' cradles, comically fluttering 
their tails, and chirping so musically that the wearied men were 
charmed to watch and listen. There was good shooting, too, for 
wild fowl and small game on the open veld ; and not far from 
Dutoitspan there was a large stretch of thickets and scrub where 
korhaans and paauws and partridges and plovers and hares 
abounded. The stately Kafir cranes shook their bluish gray 
plumes on the brink of the vleis, or water holes, where they 
came to drink, and were shot by the hunters who lay in wait. 
Their flesh was not unpalatable as a change from biltong, but 
the hunters who pushed toward the Vaal brought back better 
eating than cranes in their bags, wild ducks and geese and 
guinea fowl, and even a nimble springbok or queer-headed wilde- 
beest or hartebeest. The swift leopard, too, was occasionally 
shot near the river bank, and the rambling diggers found some 
fun in unearthing porcupines, or chasing a jackal with dogs, or 
lying in wait for the shy head of a meerkat to pop out of a 
hole. A string of fish, that could readily be caught in the river, 

1 "The Diamond Diggings of South Africa," Payton, 1872. " To the 
Cape for Diamonds," Frederick Boyle, 1873. 



THE GREAT WHITE CAMPS 217 



made a welcome meal ; and a run to the Vaal, with a few 
camping under the willows and long swims in the swirling 
current, was a gay streak of diversion from dusty diamond 
digging. 1 

Within the camping grounds there were always, in the day- 
time and evening, stirring scenes to attract the eye, antics of 
ponies and mules ; the passage of straining bullock teams with 
carts piled high with diamond-bearing ground, or wagons loaded 
with country produce ; the rambling of pedlers with packs 
and trays ; the groups of native tribesmen in trappings of skins 
and feathers or comical old clothing, chattering or singing or 
whooping or dancing ; the clustering of black women, washing 
on the edge of the water pools ; the rallies of amateur minstrels 
and travelling shows ; the merry-go-round, ridden by laughing 
children or solemn country clowns ; the rattling of pins in the 
bowling alleys and clicking of balls on pool tables, the crowded 
" canteens," and all the other lively sights and sounds of the 
fermenting camps. 

Fortune-tellers and wizards, who professed to be able to see 
through the earth, did not need to dig diamonds. Credulous 
prospectors filled their laps with silver and gold. Payton tells 
of one whose tent in Dutoitspan was thronged, day and night, 
with eager dupes, showering shillings upon her, and her income 
was reckoned at ^30 a day. Many of the Boers had unshak- 
able faith in her. When she told a poor Dutchman that there 
was nothing in his claim, he could not be persuaded to dig any 
longer. If she promised diamonds and none were found, she 
had an easy defence : " The niggers must have stolen them." 
Then the wrathful claim-holder would pounce on his Kafirs and 
haul them to the " tronk," the police station of the camp, where 
the blacks were searched to the tips of their toes. Sometimes the 
wizard guessed right, for diamond stealing was common, and 
precious stones would be brought to light with joy to the owner 
and credit to the fortune-teller. 2 

1 "The Diamond Diggings of South Africa," Payton, 1872. 

2 Ibid. 



218 THE DIAMOND MINES OF SOUTH AFRICA 

The whipping-post was soon set up in the new camps and 
freely used to chastise theft and other offences. The whipping 
was not very severe, and it was remarked that the " cat " was not 
as heavy as the one at Klip-drift. Many of the natives would 
steal anything that they could carry off, and put on a brazen 
face that would impose on a police court judge. A story is told 
of two Hottentots who took the sheepskins off a man's bed and 
tried to sell them back to their owner as soon as his eyes were 
open in the morning. They took fifteen lashes on their bare 




Natives resting, on their Way to the Mines. 

backs without a whimper. Small fines were imposed for slighter 
breaches of the camp regulations, and roaring drunkards were 
occasionally clapped into the " tronk," a weak little jail, but the 
whipping-post was necessarily the main dependence for punish- 
ment of natives. 

Strangest and most interesting of all features of the camps to 
a newcomer were the habits and antics of the marvellous collec- 
tion of savages, streaming into the Fields from the heart of 
Africa. No mining camp on earth before ever held such a mot- 
ley swarm of every dusky shade, in antelope skins and leopard 
skins and jackal skins and bare skins, with girdles and armlets 



THE GREAT WHITE CAMPS 219 

of white ox-tails, and black crane plumes and gorgeous bird 
feathers, and dirty loin cloths, and ragged breeches, and battered 
hats and tattered coats. With and without the fire of rum they 
might dash off at any moment into some wildly whirling reel 
or savage dance, gabbling in a hundred dialects, whooping with 
weird cries, and chanting plaintive, gay, and passionate strains, 
now dissonant, now sweet. Whenever a new party of "raw" 
natives came in from the wilderness, weary, grimy, hungry, shy, 
trailing along sometimes with bleeding feet and hanging heads, 
and bodies staggering with faintness, a howl of jeers was a 
common greeting, and a pelting with rotten fruits and stones 
was likely to follow the scared troop up the street of the camp, 
though the natives were not churlish at heart, and might, after- 
ward, share their last crust with the strangers. 

Their savage habits clung to them long in camp. Some 
delighted to smoke in the old native way, by making a little 
funnel in the wet ground with a slender stick and sucking the 
smoke through one end while the tobacco leaves burned in a 
hollow at the other. As a rule all the natives from Delagoa 
Bay and districts to the north of that part smoked cigars with 
the fire end in their mouths. When sheep or bullocks were 
killed at market, the natives hung about and returned exulting 
if the obliging butchers gave them the entrails to hang in fes- 
toons about their necks and carry off smeared with filth. They 
fed content day after day on a few handfuls of mealies or ground 
maize with an occasional chunk of refuse meat. They had 
little use for water except to drink, and they much preferred 
Cape brandy. After working all day, and roving about and sing- 
ing at night, they could sleep as soon as their heads touched the 
ground, on the bare earth, without shelter, or in a squalid hut 
with a dirty sheepskin wrapped around them. These quaint 
Africans, mingling in a kaleidoscopic show with adventurers 
coming from the ends of the earth, made a unique, moving 
drama on the stage of the Diamond Fields that cannot be 
forgotten by any spectator. 




CHAPTER VIII 

OPENING THE CRATERS 

N ever present danger hung over the miners 
from the very outset of their pit digging in the 
diamond-bearing funnels. The yellow ground 
was a breccia so loose and friable that it was 
constantly caving in upon the heads of the 
diggers. Then the pits were sunk so close 
together that the walls gave way and slipped, crumbling into the 
claims below. A loaded cart, passing along the edge of a road, 
would often topple over and sometimes plunge with driver and 
mule into the pit below. 

Prospecting on the Alexandersfontein farm was not long con- 
tinued ; but the diggings at Dutoitspan, Bultfontein, De Beers, 
and Kimberley were ardently opened by swarms of diamond 
seekers. The surface area covered by claims was very much 
larger than the diamond-yielding ground, whose total extent 
was, approximately, seventy acres. When the claims were con- 
solidated by purchase, many years later, the Kimberley open 
mine surface was figured to be 33 acres; De Beers, 22 acres; 
Dutoitspan, 45 acres; and Bultfontein, 36 acres. These meas- 
urements more than cover the extent of the original locations, 
which were as follows: Kimberley, 470 claims, equal to 10.37 
acres; De Beers, 622 claims, equal to 13.72 acres; Dutoitspan, 
1441 claims, equal to 31.79 acres; and Bultfontein 1067 claims, 
equal to 23.54 acres. Only a few scattered diamonds were 
found outside of the rim of "reef" enclosing the diamond- 
bearing craters. 

To present clearly the progress of mining in the several 
funnels, it is desirable to trace the advance of each separately 



OPENING THE CRATERS 



221 



through the period of the open mine working, to show the dif- 
ferent methods employed, and how one mine profited by the 
costly experience of another. The superior richness of the 
diamond-bearing ground in Kimberley mine urged forward its 
opening more rapidly than the development of the others, 
and this may properly be outlined first. The plan of min- 
ing, with the reservation of roadways determined by the Free 
State inspectors, proved a poor makeshift at best, before the 
sinking of claims had progressed many feet below the sur- 




The Breaking-up of the Roads, Kimberley Mine, 1872. 

face. The bordering-claim owners undercut the roadways cross- 
ing the mine, in working to the bounds of their allotments, 
and these reserved roads soon began to cave away in places to 
an extent that made the passage of carts very risky. It was 
doubtless convenient to have ready access to every part of the 
surface of the mine, and it was a moving spectacle when four- 
teen parallel roadways were covered with files of plunging mules 
and rumbling carts, goaded by the cries and whips of many hun- 
dreds of half-naked Kafirs or white drivers ; but it was a piti- 
ful burlesque of mining when the roadways cracked and crumbled, 
and crevasses were bridged with sliding planks, and mule carts 



222 THE DIAMOND MINES OF SOUTH AFRICA 




Miners going to Work. 



and men staggered and slipped over the roadsides into abysses. 
Yet in spite of all risks and accidents, the roads were patched 
up and maintained in some shape long after it was evident that 
they were doomed. At length no possible patching and bridg- 



OPENING THE CRATERS 



223 




The Hand Drums used for Winding-up the Blue Ground. 

ing could arrest their fall. One after another, before the end 

of the year 1 872, had crumbled away and slipped into the great pit. 

The mine was then an open, oval quarry, about a thousand 




De Beers Mine, 1873. 



feet in length and six hundred feet in extreme width. The 
broken blue ground on the face of the rough jumble of terraces 



124 THE DIAMOND MINES OF SOUTH AFRICA 

had been hoisted to the surface usually in buckets, by means of 
a rope passing around a windlass and through a pulley fixed 
in a pole set near the edge of the claim, but in 1872 a simple 
device of haulage over two grooved wheels was largely intro- 
duced. One wheel was set on the pit bottom, and the other on 
the surface, with a handle attached by means of which one or 
more stout natives could wind up a rope passing from wheel 
to wheel, carrying up a loaded bucket and lowering an empty 




Kimberley Mine, 1873. 

one. This crude device served the purpose as long as a wheel 
could be set near the edge of a claim on unbroken ground, or 
along the roadway ; but when all the claims were at the bottom 
of one huge open pit, it was obvious that only the outer tier of 
claims could be worked by this method. 

Then a haulage system of really remarkable ingenuity was 
contrived. A massive timber staging was set completely around 
the mouth of the mine, carrying two, and in parts three, plat- 
forms, one above the other. The upper platform was connected 
by strongly anchored ropes with the claims in the middle of the 
mine, and the lower platform in the same way, with the claims 



OPENING THE CRATERS 



225 




Kimberk-y Mint-, 1874. 

nearer the margin of the mines. Where there was a third inter- 
vening platform, ropes were stretched to claims lying between 
the outer and inner circle. Windlasses were set on the plat- 




Anotlier View of Kimlierley Mine, 1874. 



226 THE DIAMOND MINES OF SOUTH AFRICA 




Natives carrying Ground out of Dutoitspan Mine in Buckets. 

forms, together with guide wheels over which hauling ropes 
passed, dragging the buckets swiftly from the bottom of the 
mine on little overhead runners, rattling over the stationary 
roped inclines. When the loaded buckets reached the platform 
levels they were dumped into chutes carrying the contents into 
bags, which were readily carted away to level depositing grounds, 
or "floors," as they were technically termed, where the blue 
ground was sifted and sorted. The empty buckets were easily 
W 




Back View of the Staging with Grooved Wheels, at Kimberley. 



OPENING THE CRATERS 



227 



returned, running back by force of gravity over the ropes to 
the claims. The buckets were of rawhide, for this material 
was found to be more lasting than iron, and the ropes were at 
first largely of hemp or twisted rawhide ; but iron and steel 
wire gradually replaced all other material. 

So thickly together were these lines set, that the whole face 
of the vast pit seemed to be covered by a monstrous cobweb, 
shining in the moonlight as if every filament was a silver strand. 
Never has any eye seen such a marvellous show of mining as 




Kimberley Mine, 1875. 

was given in this grand amphitheatre, when the huge pit was sunk 
far below the surface level ; when the encircling wreath of the 
chasm rose sheer and black like the walls of a deep, gloomy 
canyon, or the swelling round of a demon's caldron ; when a 
downward glance from the perch of a platform made weak heads 
reel ; when thousands of half-naked men, dwarfed to pygmy 
size, were scratching the face of the pit with their puny picks 
like burrowing gnomes ; when thousands more, all grimy and 
sweating and odorous, were swarming around the pit's mouth, 
dragging up loads of diamond-sprinkled ground and carrying 
off their precious sacks ; when hide buckets were flying like 



228 THE DIAMOND MINES OF SOUTH AFRICA 




Snow in Kimberley Mine, June 21, 1876. 

shuttles in a loom up and down the vast warp of wires, twanging 
like dissonant harp-strings, with a deafening din of rattling 
wheels and falling ground ; and where every beholder was won- 
der-struck at the thought that this weird creation in the heart of 




Method of Hauling, De Beers Mine, 1873. 



OPENING THE CRATERS 



229 



South Africa had been evolved by men for the sake of a few 
buckets of tiny white crystals to adorn the heads and hands of 
fanciful women. 1 




The First Horse Whim, Kimberley Mine, 



With the deepening of the mine, " horse whims," first intro- 
duced in 1874, were gradually substituted for hand tackle in 
hoisting and lowering the buckets, which were enlarged tubs 




Hauling Gear and Jumpers, 



1 When Lord Randolph Churchill visited the diamond fields, while looking at a 
huge parcel of diamonds he remarked, " All for the vanity of woman." A lady, 
who heard the remark, added, " and the depravity of man." 



230 THE DIAMOND MINES OF SOUTH AFRICA 




A Nook in Kimberley Mine, 1874. 



holding five or six cubic feet of blue ground. These whims 
were timber wheels from fourteen to eighteen feet in diameter, 
set near the edge of the mine, to revolve horizontally about 



OPENING THE CRATERS 



231 



eight feet above the surface level. To turn the whim an iron 
hoop, projecting from the wheel, was attached to the harness of 
a horse or mule. The hauling rope was wound above the hol- 
lowed rim of the wheel, and each end of the rope was fastened 
to a tub, one hauling up the load of blue ground, and the other 
lowering the empty tub. 

In the following year, 1875, tne ^ rst steam winding engine 
employed at the mines was transported to Kimberley to take 
the place of horse power in moving the whim, and the first 




The Horse Whims, Kimberley Mine, 1875. 

application of modern mining methods to the South African Dia- 
mond Fields was made. This seemingly tardy development was 
due less to a lack of enterprise than to the heavy charges of 
freight transportation from the coast, ranging for years over 
^30 per ton, and to the scarcity and cost of fuel, combined with 
the lack of any positive assurance of the continuance in depth 
of the diamond-bearing ground. Such a deposit of diamonds 
as had been uncovered in the South African farm lands had 
never been opened before, and the erection of costly machinery 
for its extraction was naturally deemed an unwarranted risk. 



232 THE DIAMOND MINES OF SOUTH AFRICA 




Hauling Gear, Dutoitspan Mine, 1876. 

But as the cutting passed farther and farther down through 
the reef-circled funnels without disclosing any barren stratum or 
break in the body of breccia, the surmise rose gradually to the 
point of conviction that the funnels were craters of extinct vol- 
canoes, filled by successive eruptions of steam or gas under great 
pressure with a diamantiferous breccia, carrying fragments of vol- 
canic and sedimentary rocks and crystals of many kinds of min- 
erals. This conclusion, however, was hardly more than one of 
several varying assumptions in advance of the thorough re- 
searches and analyses of later years, when the prosecution of 
deep mining works determined positively the existence of craters, 
the character of the breccia, and the composition of its encasing 
reef. So the progress of mining on the Diamond Fields was 
long a hesitating and tentative advance, groping step by step 
into the depths of the blue ground. 

After the device of staging and hoisting ropes had solved, 
for a time, the problem of open excavation in the Kimberley 
mine crater, and the caving of the blue ground was no longer 
a terror to the diggers, the collection of water in the pits was a 
serious annoyance. Most of this water was surface drainage, 



OPENING THE CRATERS 



233 




Surface Loading Boxes. 




Aerial Trams and Surface Chutes, De Beers Mine, 18 



234 THE DIAMOND MINES OF SOUTH AFRICA 

flooding the lower levels in the rainy season, but never sufficient 
in quantity to have been any considerable obstacle, if the mine 
had been equipped with the ordinary pumping machinery erected 
in other mining districts. The lack of any such machinery, 
compelling for years the bailing and hoisting of the water in 
buckets or tubs by hand or horse power, was no slight draw- 
back to the progress of sinking. Hard upon this impediment 
came the much graver trouble occasioned by the crumbling, 
cracking, sliding, and falling of the encasing reef of decom- 
posed basalt and shale. The unstable walls of these soft rocks 
caved rapidly upon exposure to air and moisture into the open 
pit, and the fracturing and slipping were aggravated by the 
imprudent vertical cutting of the mine, removing the entire 
body of blue ground without cutting away the reef in compara- 
tively stable terraces or slopes. Obviously no single claim-holder 
would undertake the cost of removing the dangerous reef for 
the common benefit, and it was difficult to secure the general 
cooperation and subscriptions so urgently required for this work. 
What is everybody's business in theory has too often been 
nobody's business in practice. The mean and short-sighted 




Hauling Gear, Kimberley Mine, 1885. 



OPENING THE CRATERS 



235 



hope to be protected without cost by the enterprise of the more 
liberal and prudent ! The central claim holders counted on the 
distance of their claims from the reef to assure their safety, and 
the outer circles of claim-holders hung upon luck to shield their 
ground. But the frequent recurrence of reef falls and slides, 




The French Company's Sling Gear, 1885. 

together with the gathering of troublesome water pools, so 
emphasized the necessity of combination that a Mining Board 
was organized in 1874 by general concurrence of the claim- 
holders, with power to levy a comprehensive assessment to cover 
the expense of keeping the mine clear of water and fallen reef. 
This board took the place of the original " Diggers' Committee " 
which had hitherto been charged with the execution of the crude 
code of mining regulations. 

The creation of this new administrative board was a move 
in the right direction, but unfortunately it did not go far 
enough. The opening of so large a number of small separate 
claims by individual holders barred the essential application of 
system to the sinking of the great pit. The Mining Board 
lacked the means, if it had the foresight, to undertake the 



236 THE DIAMOND MINES OF SOUTH AFRICA 



checking of the reef slides by cutting back the vertical reef 
walls, and it attempted little practically besides the removal of 
the drainage and spring water and the clearing away of fallen 
reef from the face of the blue ground. This was slipshod 
mining at best, for the bare extraction of the reef, which had 
slid and fallen over the claims, actually exposed the mine to 
further reef slides, and this disaster was aggravated by the utter 

lack of system in 
clearing off the 
fallen debris. 
Every claim- 
holder was per- 
mitted to clear 
off his own claim 
independently, 
and credited with 
an allowance of 
45. for every load 
of 1 6 cubic feet 
of broken reef 
removed. The 
clearing of the 
face of one claim 
or a cluster of 
claims was no 
security against 
repeated reef slides, and barred the possibility of developing 
any section of a mine in an economical and well-planned way. 

The practical impossibility of opening a little claim, whose 
surface area was only 961 square feet, beyond a limited depth 
forced the consolidation of claims in spite of the original pro- 
hibition of "claim blocking." The poorer sections of ground 
were the first to feel the pressure for the enlargement of hold- 
ings, and, to secure the continuance of working, permission was 
granted in 1874 by the Kimberley Mining Board for the hold- 
ing of ten claims by a single owner. This concession led to 




Loading Tubs at Bottom of Kitnberlev Mine, 18 



OPENING THE CRATERS 



237 



further combination and consolidation of claims in the hands 
of partners and stock companies, but the comprehensive union 
essential to the proper development of the mine was far too long 
delayed. The mining operations of a number of individual 
claim-owners, firms, and companies whether in keen rivalry 
or in varying degrees of energy and listlessness without any sus- 




BHH 

The Standard Company's Claim, Bottom of Kimberley Mine, 1885. 

tained concert of views and means could not be prosecuted 
efficiently and prudently within the small area of a diamond- 
mine crater. Unluckily for the advance of diamond mining 
and the fortune of many struggling claim-holders, this irresist- 
ible conclusion was not made clear to the mass of miners until 
it was demonstrated after long years of costly fumbling in the 
diamond-bearing funnels. 

In view of the subdivision of ownership, the incoherence of 
the mining operations, and the lack of essential funds, the 
Mining Board can hardly be charged with a great part of the 



238 THE DIAMOND MINES OF SOUTH AFRICA 




Bottom of Kimberley Open Workings. 

burden of responsibility for the failure to save the mine from 
disaster through reef falls. The open pit working was not its 
design, but the inexperienced undertaking of a mass of diggers 
who could not be prevented from extracting the diamond-bear- 
ing ground in their own rude way. They scooped out the 
crater to a depth that made reef falls inevitable, and pushed on 
their cuts through the body of blue ground, in spite of all warn- 
ing falls and slides, long after it was apparent to any mining 
engineer that the open pit sinking could not be continued with 
safety to the workers or with profit to the owners. 

But it is impossible to approve the relief measures of the 
Mining Board. It could only check the reef falls at best, tem- 
porarily and partially, but it failed to do even this. It set up 
expensive hoisting machinery on the surface level at opposite 
ends of the mine, and sunk a large vertical shaft (Kendric shaft) 
in the reef at a point two hundred yards from the northeast edge 
of the crater, with the apparent intention of removing reef rock 
through this opening or determining the continuance of the blue 



OPENING THE CRATERS 



2 39 



ground by a drift to the crater below the pit bottom. The shaft 
was driven down to the depth of 286 feet, when a stratum of vol- 
canic rock was reached, so hard that the work was abandoned. 
No use whatever was made of this costly shaft, and no consid- 
erable attempt was made to cut back the dangerous reef wall. 
Even with the stinted means at the command of the Board, 
something might have been done to preserve the mine, and an 
energetic and well-directed push to this end would have com- 
manded at least the confidence and support of the more in- 
telligent claim-holders. So, when the caving of the reef cast 
enormous heaps of debris upon the claims in the pit, the lack of 
foresight of the Mining Board was discreditably apparent. The 
cost of removing the reef rock was then vastly increased, and 
the burden was the heavier because the reef falls prevented the 
extraction of the buried blue ground. 

Two of the larger companies, the French and the Central, 
holding claims in the mine, were the first to undertake the re- 
moval of the solid reef on any extensive scale, by sinking shafts, 
in 1878 1879, at points several hundred feet distant from the 




Pumping Engine in Kimberley Mine, 1875. 



440 THE DIAMOND MINES OF SOUTH AFRICA 

north and south sides of the mine. By this means considerable 
reef was removed, and a third shaft was sunk in 1882 through 
the northeast reef border to check the imminent peril at that 
edge of the mine. To supplement the service of these shafts 
inclined tramways were opened on the west and east sides of 
the mine to cut back the upper reef walls, while wire tramways 
were stretched from the bottom of the mine to the surface edge 
to carry off the fallen reef in large tipping tubs, holding from 
1 6 to 32 cubic feet of broken rock. At the end of 1881 tram- 
ways, aggregating 19 miles in length, had been constructed by 




Incline Tramway for Hauling Reef, 1878. 

the claim-holders and the Mining Board. Steam pumping 
engines had been put in to pump out the influx of water, and 
this obstacle was, at last, easily overcome. To hasten and 
cheapen the extraction of blue ground, drilling and blasting were 
substituted for hand labor with picks, and the work of mining 
was pressed with incessant energy. But the sliding, falling reef 
mocked every effort to withstand it. 

The work of removal was undertaken too late. The reef 
slipped faster than the tram cars and tubs could haul it out. 
In 1878 more than a quarter of the surface of the claims in 
the mine was covered by fallen reef. The cost of removal, 
at the original allowance rate of 4^. per load of 16 cubic feet, 



OPENING THE CRATERS 



241 



mounted so high that the Mining Board was constrained to cut 
down the allowance to is. 6d. y but even with the rate reduced 
the expenditure for reef work and drainage in 1879 and 1880 
ran over ^1 50,000 a year, and in 1881 it rose to over ^200,000. 
Still, the need of stimulating extraordinary exertion was then so 
apparent that the rate was put up to ys. yd. a load in October, 
1 88 1, and for the eighteen months following fifty-six million 
cubic feet of broken reef were hauled away by the claim-holders 




Hauling Reef, Kimberley Mine, 1873. 

alone, at a cost to the Board of over ^650,000, without reckon- 
ing the amount extracted by the operation of its own tramways. 
This stupendous charge was obviously too heavy to be 
borne even by the richest diamond mine, and no assessment 
scheme could sustain it. The Board struggled for months 
under the load, issuing notes when it had no cash in hand ; but 
in March, 1883, its issue of outstanding notes or " reef-bills " was 
so great that its book showed a debit balance of over ^2 50,000, 
and the local banks would extend no further credit. The Board 
was bankrupt, reef extraction was stopped, perforce, and the 



242 THE DIAMOND MINES OF SOUTH AFRICA 



claim-holders were face to face with an appalling situation ; for 
in spite of all efforts and the outflow of money like a water- 
spout, the resistless reef was unchecked. The mine walls con- 
tinued to fall in faster than they could be hauled out, and even 
central claims in the mine were buried. The gloomiest forebod- 
ings fell like a black cloud on the spirits of claim-holders. In 

the judgment of 
many observers, the 
great Kimberley dia- 
mond mine was 
doomed beyond hope 
of resurrection. 

The open pit had 
been sunk to the 
depth of something 
over four hundred 
feet, in the lowest 
working, at the end 
of the year 1882. In 
order to haul out one 
million loads of blue 
ground during that 
year, three million 
loads of reef had 
been raised. The 
cost of hauling was 
increasing with the 

Reef Falls, Kimberley Mine, 1881. deepening of the 

mine, and owing to the reef falls, the production of diamonds 
was disastrously sinking. In 1883 tne ' ac ' c f funds only per- 
mitted the lifting of one and a half million loads of reef at a cost 
of ^250,000, and the output of blue ground sunk to 350,000 
loads. In November of that year a long portended reef slide 
cast 250,000 cubic yards of shale upon the face of the pit, piling 
its mass on the claims half across the mine. This was seemingly 
a crushing infliction. It was, at least, a conclusive proof that 




OPENING THE CRATERS 



243 




244 THE DIAMOND MINES OF SOUTH AFRICA 

open pit sinking was no longer feasible even for the richest 
claim-holders. About four million cubic yards of reef had been 
hauled at a cost of nearly ^2,000,000, yet there was no check 
to the reef falls and slides. At the close of the year the Inspec- 
tor of Mines reported 
that " only about fifty 
claims had been regularly 
worked during the past 
year." The field for the 
operation of individual 
claim-holders was deci- 
sively closed. The only 
hope for the mine was in 
the prosecution of deep 
and extensive under- 
ground works by the 
combination of claims in 
hands able to conduct 
such operations success- 
fully. 

In advance of such 
an undertaking the yield 
of the mine was fortu- 
nately sustained by an 

The Central Company's Shaft, Kimberley Mine, 1885. CXDCrt makeshift. IVIr. 

Edward Jones, a trained mining engineer, had been one of the 
leading contractors for the removal of reef, and had given close 
study to the problem of the continuance of the extraction of blue 
ground. Through his design and insistent confidence, in spite of 
all doubts and sneers, a shaft was sunk through the mass of fallen 
reef at the bottom of the deepest part of the mine by lowering a 
square timber frame and shovelling out the loose rock from the 
inside of the enclosure. The frame was constructed in sections on 
the plan of a coffer dam, adding section to section from the top until 
a stout timber shaft passed entirely through the broken shale and 
entered the underlying blue ground. The shaft was then read- 




OPENING THE CRATERS 



245 



ily extended, and drifts from this opening were made through 
the blue ground. The peculiar service of this device was its 
saving of hundreds of feet of costly shaft cutting through the 
solid reef to reach the blue ground a very desirable contribu- 
tion at a time when the richest claim-holders were sharply pinched 
by the failing mine and the discouragement of capital. The cost 
of all development work was defrayed by the blue ground 
extracted in opening the drifts and cross-cuts, so that there was 
no further delay in resuming operations in the mine. The first 
shaft had been sunk on the ground owned by the Central Com- 
pany, and it was soon copied by a number of similar shafts in 
other parts of the mine. This brought about a most welcome 



s 




The Bottom of Kimberley Mine, 1885. 



revival of mining, and was so far highly beneficial to the labor- 
ers, claim-owners, and townspeople of Kimberley, though it was 
not designed for permanent service. 

While the blue ground was being removed through shafts 
sunk in the bottom of the open mine, it was apparent to all that 



246 THE DIAMOND MINES OF SOUTH AFRICA 

the life of these shafts must be very short. Preparation was 
therefore made for future work by sinking shafts outside the 
margin of the open mine, and at sufficient distance from it to 
insure them against any probable caving of the surface ground 
in their vicinity. Vertical shafts were sunk by the Central and 
French companies, and tunnels driven from them. The plan 
of Kimberley mine, 1883, shows these tunnels. 



.WMBERUr 




Before describing the subsequent application of engineering 
science to underground mining, it is desirable to trace the prog- 
ress of the other mines on the fields to the period in develop- 
ment reached by the leader. The claim-owners in De Beers 
mine profited greatly by the object lessons given in the opening 
of the great pit of Kimberley. For the first twelve years after 
the discovery of the mines, the Kimberley mine ran far ahead of 
the others from the superiority of its yield for some distance 
below the surface. The fatal error of the neglect of the claim- 



OPENING THE CRATERS 



247 




Reef Slips, Kimberley Mine, 1874. 




Kimberley Mine, showing how the Ground cracked before Subsidence. 



248 THE DIAMOND MINES OF SOUTH AFRICA 



owners and Mining Board to cut back the mine walls was appar- 
ent in time to save many thousands of pounds to De Beers. 
This mine was also fortunate in the comparative hardness and 
stability of the basaltic rock stratum overlying the shale and 
forming the marginal top of its pit walls. By cutting back the 
reef in terraces, the De Beers Mining Board saved the mine 



from any serious rock falls for a con- 
years. Only two hundred and fifteen 
of reef were removed in the five 
1882, but this sufficed to 
the time. The cost of 



siderable number of 
thousand cubic yards 
years ending with 
protect the mine for 
its removal was only 
slight burden com- 
the charges at Kim- 
mine, and showing a 




The Central Compan 



cost per yard or per load of reef raised much less than the Kim- 
berley average. This was a signal demonstration of the advan- 
tage of prudently cutting away the reef before it fell into the pit 
and buried prolific claims and increased the hauling charges. 

This precaution, however, did not suffice to shield the mine 
from disaster when the pit was greatly deepened after the reef 
falls at Kimberley had diverted mining enterprise to De Beers. 
Over one hundred and forty thousand cubic yards of solid 
and broken reef were removed in 1883 and 1884, but reef slides 
were fast increasing, and it was judged necessary by the Min- 
ing Board to stop any further outlay for reef hauling when the 
mine bottom was 350 feet below the surface. The diamond- 
bearing ground had then been scooped out of the larger part of 



OPENING THE CRATERS 



249 




The Last of Open Working, Kimberley Mine, 1889. 

the funnel, but there was still a large area of yellow ground at 
the west end which had not yet been extracted because it con- 
tained so few diamonds compared with the other parts of the 
mine. The falls of reef had covered the eastern end of the 



250 THE DIAMOND MINES OF SOUTH AFRICA 



mine, and early in 1885 the west end yellow ground caved in, 
and an enormous mass of nearly five million cubic feet fell in 
one day to the bottom of the mine, overlapping the fallen reef 
and burying the claims still open for work. This disastrous fall 
forced the stoppage of mining for six months until some part of 
the reef and yellow ground could be taken out, and mining was 
then resumed in a partial and half-hearted way in the open pit, 
though it was evident that further pit sinking in the face of such 

disasters was irrational mining. 

The only possible resource 
was the introduction of a system 
of underground mining, and the 
first attempt in this direction was 
made in 1884 by the opening of 
a large circular shaft at a point 
1000 feet from the north margin 
of the mine. This shaft was sunk 
vertically about 320 feet in the 
reef and then abandoned as too 
costly. In its place an incline 
was sunk, starting from a point 
about 150 feet from the west 
side of the claims, and entering 
the mine at the edge of the amyg- 
daloidal trap underlying the basalt 
and shale, so as to avoid the expense of cutting through this 
hard rock. This work was begun none too soon, for before the 
end of the year 1887 further open pit working was proved to 
be utterly impracticable, and was wholly abandoned when the 
deepest open digging had been carried in three years only fifty 
feet farther than the depth of 350 feet reached in 1884. 

Dutoitspan mine opening was practically the same as the 
course followed in Kimberley and De Beers. Owing to the com- 
parative poorness of the diamond-bearing ground, pit sinking 
was not pushed as rapidly as it was at Kimberley, and, in 1874, 
most of the miners went over to Kimberley and were glad of the 




R. D. Atkins. (Manager of Kimberley 
Mine in the earlier days.) 



OPENING THE CRATERS 



251 




No. 2 Incline Shaft, De Beers Mine. 

chance of working over the " waste ground " which had been 
cast away from the cradles and sieves of the early diggers. Two 
years later, when improved methods of handling the ground 
were coming into use, the miners flocked back to the abandoned 
ground and took out fresh claims. Warned by the experience 
of Kimberley, a circle of solid blue ground was left as a buttress 
against slides and falls of the encasing reef of shale, and for ten 
years this expedient served to shield the miners. 

But this safeguard failed when the open working had reached 
a few hundred feet in depth. Warning surface cracks had been 




Eldorado Road, Dutoitspan Mine, 1874. 



252 THE DIAMOND MINES OF SOUTH AFRICA 

noticed on the northern margin of the mine, but the ardent 
diamond seekers kept on digging recklessly, until one day in 
March, 1886, when a huge mass of blue ground and reef broke 
away suddenly from the northern end of the mine and rolled 
over like the surge of a monstrous breaker, falling hundreds of 
feet with a fearful crash upon the doomed men at the bottom of 




Claims in Dutoiispan Mine. 

the pit. The loss of life would have been frightful, but happily 
for the miners the fall was at the noon dinner hour, when the 
work of hoisting blue ground was stopped and blasting in the 
mine was begun. Most of the workmen had left the mine, but 
eighteen poor fellows eight white men and ten Kafirs had 
taken shelter from the blasting in a pumping engine house in 
the pit. The avalanche of rock fell on the house, and every one 
in it was fatally crushed or scalded by the escaping steam. One 
hundred thousand cubic yards of shale and blue ground buried 
the claims on the pit bottom, and this fall was followed by others 
which ruined the open workings in 1887, when the mine had 
reached a depth of 400 feet. 

In Bultfontein there was only another variation of the same 
tale of open pit working and final wreck. The work of extract- 
ing the yellow and blue ground was well planned at the outset, 
under existing circumstances, by the cutting of inclined road- 
ways over which the ground was hauled in bullock carts. In 



OPENING THE CRATERS 



253 



1880 effective hauling machinery was substituted for the carts, 
and the precious ground was extracted so rapidly that the depth 
of about five hundred feet was reached in the open working, a 
point probably beyond any attained in the other pits. Here, 
too, as at De Beers, there was an effort to protect the mine by 
cutting back the reef in terraces ; but this safeguard was tried 
too late, and in any event it could only have deferred for a few 
years the fate of the mine. Before the close of the year 1889 
almost the whole of the pit bottom was covered with fallen reef 
and only four engines were at work hauling blue ground. 

The Extraction of the Diamonds 

While the sinking of the pits was progressing with improved 
mining appliances, there had been a considerable advance in the 
methods of concentrating the diamond-bearing ground and win- 




Bultfontein Mine, 1879. 

ning the diamonds. For the first three years after the opening 
of the mines, the handling of precious ground was exceedingly 
crude and wasteful. The broken ground taken from the craters 
was crushed more or less finely by pounding with shovels and 



154 THE DIAMOND MINES OF SOUTH AFRICA 

mallets and clubs. Then it was sifted in rocking troughs, fitted 
with sieves like the placer miners' cradles, and the concentrate of 
pebbles and crystals and coarse rock grains was spread on tables, 
or sheets of iron and wood laid on the ground, where it was 
scraped over by hand, and the gems picked out. In this rough 
process a third and perhaps a half of the smaller crystals were left 
in the waste ground, and the losses from theft were enormous. 

In 1874 there was a change for the better in the introduc- 
tion of water in concentrating. By building dams and sinking 




The First Rotary Washing Machine. 

wells the water supply of the camp was increased materially, and 
it was possible to divert a portion for the diamond-washing 
appliances. Most of the early machines for this purpose were 
simple cradles with riffles or ridges set at intervals on the 
bottom, and a sieve at the end. The pulverized ground was 
dumped into a cradle with a sufficient flow of water to carry off 
the slime, while the rocking shook the ground, and caused a 
settling of the heavier mineral deposit at the bottom. With one 
of these rockers from six to thirteen cartloads of ground were 
washed in a day. Another device was a circular trough or pan, 



OPENING THE CRATERS 



255 



fitted with a revolving set of iron teeth like a comb, that stirred 
the ground and water and caused the settling of the concentrate. 




Another Early Washing Machine, 1874. 

This puddling trough would concentrate from twenty-five to 
thirty-five cartloads in a working day and cost at first, about 
^250, while the simpler cradle could be bought for ^15, or 
less. There were other more elaborate devices, but their cost 
put them out of the reach of the ordinary digger. All were 
based on one adaptation or another of the puddling principle, 
and the fall and separation of minerals of different specific 
gravities. 

The sorting of the concentrate from the 
puddling troughs was done by the same 
method employed after the dry sifting, 
but there was some improvement in the 




Horse-power Washing Machine, 1875. 



256 THE DIAMOND MINES OF SOUTH AFRICA 




OPENING THE CRATERS 



257 



. 




258 THE DIAMOND MINES OF SOUTH AFRICA 



precautions against loss by theft. The natives, who were com- 
monly employed in scraping and picking over the mineral de- 
posit, were more carefully watched. Some were lodged in tents 
and sheds adjoining the stables belonging to claim-owners, and 
there was some oversight of them by night as well as by day. 
When the claim-owners combined in companies, their workmen 
were frequently kept together in enclosures called "compounds," 
where they were furnished with food and shelter at moderate 
charges deducted from their pay. This sepa- 
ration and partial restriction was of undoubted 
service, not only in diminishing the oppor- 
tunities for successful theft and disposal 




Washing Gear, Bultfontein Mine. 

of stolen diamonds, but in checking the drunkenness of the black 
workmen and the outbreaks in the canteens and streets. 

Progress was made, too, though much too slowly, in the 
more perfect pulverization of the blue ground. It was soon 
observed that the broken ground would crumble upon exposure 
to the air, and after some weeks or months, according to its 
hardness, a mass of breccia, thinly spread out and raked over, 
would be very largely decomposed to fine sand fit for washing, 
without further treatment. This natural pulverization was far 
cheaper and better than crushing with mallets ; but the burden 
of accumulating and storing great quantities of ground was too 
heavy for the ordinary claim-holder, who was dependent upon 
quick returns : so only the larger companies maintained stores 



OPENING THE CRATERS 



259 



of ground on their depositing places or " floors," and none of 
these, even, were disposed to wait for the adequate pulverization 
of the ground by the natural agencies of the sun, air, and rain. 
Still the floors were gradually enlarged on the veld, and were 
frequently fenced in with wire. Year by year an increasing 
proportion of blue ground was pulverized. The average yield 
of a truck load, or sixteen cubic feet of blue ground, from Kim- 




steam Washing Gear, Kimberley Mine. 

berley mine, was computed to be one carat in diamonds, a valu- 
ation ranging from twenty-eight to thirty-six shillings, according 
to prevailing market rates. 

The mining camps changed, year by year, more completely 
to the appearance of thriving mining towns. De Beers fused 
with De Beers New Rush in the town of Kimberley, while 
the town of Dutoitspan rose on its camp site two miles away. 
The connecting roadway was lined with straggling houses. 
There was little available timber fit for building purposes, but 
galvanized iron was very largely substituted for the canvas tents 
during the first ten years, and, from 1880 on, many brick build- 
ings were erected at Kimberley. Outside of the main business 
street there was little attempt at first, to lay out regular avenues, 



260 THE DIAMOND MINES OF SOUTH AFRICA 

and the diggers shifted their tents or " tin houses " to any 
vacant place that suited their fancy. The little galvanized iron 
buildings were so light and strongly riveted that they could be 
picked up and carried away by a few strong Kafirs. But with 




Webb's Washing Machine, 1878. 

the growth of the towns stands became more valuable, and title 
and possession were more sharply looked after. In 1876 the 
valuation of the town of Kimberley for assessment or taxation 
purposes was $5,151,500. Churches, schools, banks, hotels, 
theatres, concert rooms, and stores and offices of various kinds 
were erected to answer the demands of a prospering mining town. 
Sidewalks were laid along the principal streets, and after 1874 
there was a regular appropriation for street watering. The 
houses grew in size and stability. Verandas and porticoes were 
added in place of the roof projections that gave a little shade to 
the early diggers, and many of the dwellings were set with a 
fringe of garden in front or on the sides, in which fruit trees and 
vines and choice flowers were planted. 

With the advance of the diggings in depth, the combination 



OPENING THE CRATERS 



261 



of claims, and the ending of widespread prospecting, the influx of 
whites to the camps fell off greatly. The shifting population 
of prospectors dropped to the number that could find employ- 
ment in the mines or in the dependent towns. It was estimated 
in 1876 that the white population of Kimberley was about eight 
thousand, and the native from twelve to fifteen thousand. In 
Dutoitspan and Bultfontein there were perhaps six thousand 
more of whites and blacks. 

The character of this population has been most absurdly 
decried. " The Diamond Fields of South Africa," writes one 
flighty reporter, " have been hot-beds of rowdyism. The liber- 
tines, forgers, bird-catchers, and other outcasts of Europe have 
found a refuge there as in Alsatia of old. The Houndsditch Jew 
and the London rough reign supreme." Thousands of wit- 
nesses might be summoned, if necessary, to refute this nonsense. 
Libertines and forgers drift elsewhere for prey than to hot, dusty 




Cape of Good Hope Company's Washing Gear, 1878. 

mining camps in the midst of the karoo ; though dainty folk 
might shrink from the roughness and grime of the diamond dig- 
gings, and weak nerves might be shaken by the boisterous exu- 
berance of the bustling camp, the restless crowd tramping the 
streets, the uproarious canteens and music halls, and the capers 



262 THE DIAMOND MINES OF SOUTH AFRICA 

of motley diggers and wild Africans. Liquor drinking ran to 
excess, as it always does in a prosperous mining camp, and the 
natives especially were given to drunkenness ; but the wildest 
sprees rarely threatened danger to life, for the hot spirits were 
blown off in yells, chants, and dances. Every accurate record 
shows that murder and robbery and the more flagrant and brutal 
crimes were notably rare compared with the showing of the early 
American and Australian mining camps ; and when the turbu- 
lence of the rush was over, and the bubbling camps simmered 
down to the comparative order and steadiness of the working 




Washing Gear, Dutoitspan Mine. 

mining towns, there was little disturbance from any outbreak 
of ruffianism. In spite of all demoralizing influences, the con- 
servative and civilizing agencies and public spirit that advance 
communities and exalt good citizenship gained in force year by 
year on the Diamond Fields. 

Notable progress was made in the provisions for the health 
and security of the towns. The most crying need, from the 
first, had been pure and abundant water. The average rainfall 
of the mining field was only 17.5 inches, and the suffering from 
the lack of water in the dry season was scarcely endurable. 
Much was done to improve and increase the supply by the sink- 
ing of wells and extension of natural reservoirs and the more 



OPENING THE CRATERS 



263 



general introduction of filtering appliances. Dr. Morton noted 
in 1876 a marked advance in the health of the population on 
the Fields. The death rate at Kimberley, he said, was exceed- 
ingly small. The most sickly months of the year were August 
and January, marking the effect of the extremes of cold and 
heat. Outside of the ailments incident to the dust and exposure 
and sudden variations of temperature, there was little disease, 
and he particularly observed the complete immunity of the 
field from hydrophobia, though every man, woman, and child 
appeared to have a dog at their heels. 




Washing Gear, Bultfontein Mine, 1878. 

It was soon perceived, however, that a more certain and 
sufficient supply of water must be obtained to meet the growing 
demands of the towns and mines. This was secured through 
the enterprise of the men associated in the Kimberley Water 
Works Company, by the construction of a pumping station at 
Riverton on the Vaal River and the laying of a main sixteen 
miles in length to a reservoir on a ridge of the Bultfontein 
farm, near Kimberley. The water from the river was raised in 
three stages by powerful compound condensing engines, and car- 
ried to the large reservoir on the ridge, five hundred feet above 



264 THE DIAMOND MINES OF SOUTH AFRICA 

the river level. From this reservoir it was distributed by a pipe 
and hydrant system to the towns and the mines. Since the 
construction of this fine plant, the towns have been supplied with 
filtered water at a cost of is. per 100 gallons ; and mines using 
great quantities have a concession materially lowering this rate. 
The amount of water sold to Kimberley annually has run as 
high as 230,000,000 gallons and more than 300,000,000 have 
been supplied to the mines. The cost of the machinery and plant 
was over ^300,000. Mr. E. A. Cowper, the consulting engi- 
neer of the Water Works Company, designed the machinery, and 
Mr. George Buchanan, C.E., was the constructing engineer in 
the erection of the plant. 

The maintenance of peace and order on the Diamond Fields 
was helped forward materially by the construction of " com- 
pounds," providing good lodging and food for the natives, check- 
ing their drunkenness, promoting steady industry, and enforcing 
restrictions essential to the common security. The police force 
of the towns was from the start so small that the tolera- 
tion of this condition attests the comparative rarity of brutal 
crimes on the Fields. Its very marked improvement with the 
growth of the town, in later years, was rather due to the rising 
demand for advance in every civic and social condition than to 
any increase in disorderly conduct or the commission of crimes. 

Diamond stealing and illicit diamond buying were, beyond 
all question, the worst plague of the camps and towns. Outside 
of this line of operation there was practically no opening and no 
temptation for the professional thief and receiver of stolen goods ; 
but the opportunities were unfortunately too apparent and easy 
for filching and disposing of diamonds. The sharpest oversight 
could scarcely prevent nimble-fingered workers from slyly secret- 
ing tiny crystals in picking over the concentrates on the sorting 
tables or in handling the deposit in the rockers and puddling pans. 
While the natives were allowed to rove about freely after their 
day's work was done, they had little difficulty in transferring the 
diamonds to the hands of the sharpers, who were always in wait 
for the chance of buying stolen stones for little money. 



OPENING THE CRATERS 265 

Offices were opened by diamond buyers in the mining towns, 
either as independent merchants or as representatives of large, 
foreign wholesale dealers and diamond cutters, and besides these 
established purchasers, there were a number of traders who made 
regular rounds through the diggings, buying from claim-owners 
in their tents or houses or at the sorting table. These peripa- 
tetic dealers were familiarly known as " kopje wallopers," for 
kopjes were the sites of the chief surface digging. No doubt 
there were dishonest men among these dealers, small and large ; 
for the frequent temptations were too strong for slight scruples, 
and it is certain also that many diamonds were bought under 
cover by saloon and shop keepers and other speculative traders 
who came into familiar contact with the diggers. 

It is plain that it was impossible to trace or identify a stolen 
diamond, even when the theft was known, and great quantities 
of gems were secretly bought and carried to the coast towns 
for sale or forwarded stealthily to foreign markets. It has been 
estimated that fully fifty per cent of the diamonds taken from 
the diggings in the early years were secreted and sold specula- 
tively. This is undoubtedly an extravagant reckoning, but 
there is no question that a large percentage were filched away. 

To give some idea of the enormous quantity of diamonds 
that were stolen in the early days of the fields, and before the 
compound system was adopted, the following notice is repro- 
duced : 

NOTICE 

The undermentioned rough and uncut diamonds having from time 
to time been recovered by this Department, notice is hereby given to all 
whom it may concern, that unless proof of the bona fide right to the 
possession of such diamonds be given, or a proper permit for the same 
be produced within ten days from the date hereof, such diamonds will be 
sold and the proceeds of such sale carried to the account of the Govern- 
ment. 

JOHN FRY, 

Chief of Detective Department of Griqualand West. 
MAY 24th, 1883. 



266 THE DIAMOND MINES OF SOUTH AFRICA 

Underneath the notice was a schedule showing 
The number of carats. From whom recovered. How acquired. 

The number of carats ranged from half a carat to 6375 
carats, which were found in the possession of one man. The 
total number reached 8443 carats, which were recovered from 
fifty persons. Two days later a similar notice appeared stating 
that 1573^ carats had been recovered, having been found in the 
possession of a well-known dealer in illicit diamonds. The 
total value of these two lots would amount to 30,000 or 
.40,000. 

The practice of illicit diamond buying was so persistent and 
obnoxious that it was curtly styled I. D. B., and the strictest 
possible regulations were made to check it and punish offenders. 
A Special Court was established in 1 8 So 1 to try cases of this kind, 
and a special police force formed with warrant to make the most 
rigorous search of suspected thieves and receivers. Under the 
Diamond Trade Act every parcel of diamonds taken from the 
Fields must be formally described and registered, and every 
transfer recorded from the date of discovery till the final ship- 
ment from the Cape Colony. No person was permitted to deal 
in diamonds unless he held a formal license, and his record books 
of purchase and sale were always open to police inspection. 
Thefts of diamonds and illicit purchasers were punished with all 
possible rigor. 

1 A Special Court was established under ordinance No. 8 of 1880. A barrister 
was appointed as special magistrate to act with the resident magistrate and the addi- 
tional resident magistrate. Under Act No. 48 of 1882 the special court for min- 
ing offences consisted of three persons, of whom at least one was a judge of the 
Supreme Court. The other two were usually the resident magistrate and the civil 
commissioner. By proclamation No. 144, dated September I, 1882, the districts of 
Kimberley, Herbert, Hay, and Barkly were within the jurisdiction of the Special 
Court. Act No. 34 of 1888 provided that the Special Court should consist of three 
members, two of whom must be judges of the Supreme Court. Persons convicted 
by the Special Court might appeal to the Supreme Court. 



CHAPTER IX 




THE MOVING MEN 

N the rush of adventurers over the Diamond 
Fields the individual was inevitably merged in 
the mass. He might feel the pulse of latent 
powers, the unslaked thirst of ambition, but 
he must be for the time no more than a drop 
of water in the rapid, a locust in the swarm. 
He was one of a myriad which exulted in the enforced equality 
of living and opportunity. 

There can scarcely be a purer democracy than an infant 
camp in such a field. Imperial sovereignty or feeble state asser- 
tion barely cast a shadow of authority over the stretch of " No 
Man's Land," the chrysalis of the Colony of Griqualand West. 
One man here was as good as another in his own mind, and free 
to maintain it. In the seething stream of humanity that poured 
into the Diamond Fields it mattered not whether one was to the 
manor born or cradled in a manger, the son of a peer or a beg- 
gar's brat. In the hot scramble for diamonds in the dirt, all 
ranks were levelled. The rough sailor jostled the captain, the 
university graduate swung his pick side by side with the navvy, 
and the last of the Vere de Veres snored in his sheepskin kaross 
back to back with a hopeless Japhet. The representative 
" Diggers' Committee " was merely the executive hand of the 
body of prospectors, the instrument of the will of the masses. 
The distribution of the diamond beds from the start marked 
the strain for equality, the hostility to aggrandizement ; and the 
relation, of demand to supply compelled the division into little 
patches of holdings. It was years before the acquisition of more 
than two claims by one person was tolerated, and only imperious 

267 



268 THE DIAMOND MINES OF SOUTH AFRICA 



necessity forced the further consolidation of claims when the 
mines had reached a depth that made patch-working impracticable. 
In this mass movement and equalizing of opportunity, the 
rise and display of strong individuality were necessarily subdued 
and slow to appear. In the years of the rush and the early 
advance of the mines, it is the life of the mass and not of the 
fractional unit that makes the history of the Fields. But with 
changing conditions, as the years rolled on, the way was opened 
for individual assertion, influence, and distinction. Then the 
men, hitherto unmarked, stood up preeminent. Then the brains 
that were capable of great conceptions and great performances 
found pressing occasion for all their foresight and energy. The 
history of the great mines that have explored the diamond-bear- 
ing craters so far beyond the pitfalls of the prospecting diggers 
is very essentially a story of remarkable men. 

In July, 1873, a young Hebrew, Barnett Isaacs, took passage 
from England to Cape Town at the call of his brother from the 

new Diamond Fields. His 
grandfather was a learned and 
honored rabbi, and the good 
standing of his family was 
marked by the marriage of 
his father, Isaac Isaacs, to a 
relative of Sir George Jessels, 
Master of the Rolls. But the 
son of the rabbi was only a 
small, plodding, frugal shop- 
keeper in London. His sons, 
Henry and Barnett, were 
trained in the excellent He- 
brew Free School in Spital- 
fields,but both boys left school 
at the age of fourteen to help their father in his shop. Henry 
was drawn away in the current of the early rush to the Diamond 
Fields in 1871, and had such success as a kopje walloper that he 
wrote home to urge his brother to join him. 




Barnett Isaacs. 



THE MOVING MEN 269 

To the restless spirit and purely speculative mind of Barnett 
Isaacs there was magnetic attraction in such a field with its 
novel and gleaming opportunities. With instant decision he 
took the steamer for Cape Town, and made the tiresome trip 
over veld and karoo to Kimberley with unfailing pluck and 
good temper. 

He was only twenty years old, and outwardly no more than 
a light-hearted boy, bubbling over with high spirits and comical 
conceits. But his fondness for athletic sports, theatrical extrava- 
ganzas, and practical jokes, and his contempt for conventional 
restraints, were merely the surface froth covering invincible 
energy and facile grasp of opportunities. He had an unshak- 
able self-reliance, a quick perception, and a fertile resourceful- 
ness that bore him up when feebler men sank. One could 
scarcely cast him in any society or any place on earth, where his 
nimble wits would not win him a living. 

The impulse to go ahead was in his blood. " It has always 
been a superstition with me," he said, " never to turn back." 
He grew apace with the calls upon his powers. He did not pro- 
fess to know more than he knew, but he was never content to 
know anything that interested him by report. " I must look 
into everything that concerns me for myself." This determina- 
tion was a safeguard. He once boasted, in a rare fit of parade, 
that he had never made a mistake in his investment of money 
in his life. But his incessant activity was fatally wearing. He 
could not dawdle. He could hardly rest. For many years his 
extraordinary vitality and endurance kept him running. He 
had the precious faculty of dropping off to sleep at any moment 
of relaxation, and awaking after slumbering for a few moments. 
Nevertheless no creature of flesh and blood could endure the 
strain which he bore and recklessly courted. " Some day such 
a bundle of quivering nerves must snap, either life or brain must 
go," said one of his closest friends. But when young Barnett 
Isaacs wandered into Dutoitspan, "fit for anything," as he him- 
self declared, after his long tramp and meals of porridge and 
biltong, nobody saw in him the raw material of one of the 



I 7 o THE DIAMOND MINES OF SOUTH AFRICA 

remarkable financiers of the century, or forecast, even dimly, the 
meteoric career of Barney Barnato. 

His brother Henry had fancied and taken the name of Bar- 
nato, as a professional shift from his own family name, when he 
first came to the Diamond Fields and tried his luck first as a 
conjurer and vaudeville performer, relying upon the sleight of 
hand proficiency which he had gained in boyish practice to 
amuse his friends. Henry soon turned his hand to the more 
profitable business of a diamond trader, but his stage name stuck 
to him, and passed naturally to his younger brother, who accepted 
it with easy indifference. So young Barnett Isaacs became 
familiarly known as " Barney Barnato," and for the first year 
or two of his life on the Diamond Fields floated along in the 
current as " Harry Barnato's brother." But his head never 
sank below the surface for a moment. His first buoy was 
a cigar box. He had money enough to buy sixty boxes of 
cigars after paying his way to Kimberley. With this working 
capital he went into partnership with Louis Cohen, another new- 
comer, who had started as a kopje walloper. The two young 
Hebrews picked out a shanty to their liking for an office. It 
was a little tin shed, eight feet by six, owned by an Irishman 
who offered it for rent at a guinea a day. 

" That is ridiculous," said Cohen. 

"I don't know that," said Barnato. "The situation is good, 
why not pay a guinea a day if you can make thirty shillings ? " 

This keen measuring was typical. Barney Barnato never 
counted cost alone if he wanted anything, but weighed it instantly 
against probable profit. He was never a thoughtless or reckless 
buyer. He did not shut his eyes to the risks of loss. On the 
contrary, he reckoned risks with exceptional accuracy and pre- 
cision of detail, but he reckoned profits with the same even- 
balanced judgment. Hence he was not afraid to venture when 
others shrank back. He was naturally sanguine. He had faith 
in himself, and put all his working force into everything that he 
undertook. So his high-pressure energy, persistently maintained, 
won success where a weaker and idler man would have failed. 



THE MOVING MEN 271 

There was no peculiar luck in his favor. Thousands around 
him had equal chances or better. He went to the front because 
he had the brains to choose aright and the working powers to 
make his choice profitable. He made mistakes as men of his 
sanguine temper must, but he did not make many mistakes, and 
no fatal or even greatly damaging ones. 

There is no business without risks. The most prudent man 
cannot engage in mining or in trading in mineral products with- 
out risks. If hot-headed speculation has swamped fortunes in 
such a field, it is no less certain that overstrained caution has 
failed to win anything memorable. There is a happy and rare 
mean of sagacious judgment in mining operations, and Barney 
Barnato proved his possession of such judgment incontestably. 
His mind worked so quickly, and his mental calculations were 
so exact and minute, that it was often supposed that he jumped 
at conclusions. " Barnato's snap judgment," sneered a man 
whom he outbid in competition ; " Barnato's sheer luck," 
growled the man who saw his judgment turn to gold. 

The young partners, Barnato and Cohen, worked hard, early 
and late. Barnato's keen eye gained a valuable business con- 
nection in a way that suggests his kinship to Sherlock Holmes. 

One of the most successful " kopje wallopers " (a name given 
to men who visited the various miners' huts for the purpose of 
buying diamonds) made regular rounds through the diamond 
fields on an old, lame, yellow pony, calling on men who had 
the best bargains in diamonds to offer. Barnato and Cohen 
tried repeatedly to follow him, but his track was soon lost in 
the labyrinth of tents, huts, and sand heaps. However, Barnato 
was able to see that the trader's pony had the habit of stopping 
at places where choice bargains were made, and when the broken- 
down beast was offered for sale one day by its owner, Barnato 
snapped at the chance to buy him for 17 IQJ., an enormous 
price for the old pony as a steed, but a great bargain for the 
keen diamond broker, for the walloper's business went with his 
pony, as he afterward saw to his chagrin. 

Soon Barnato became known as a " walloping walloper," 



472 THE DIAMOND MINES OF SOUTH AFRICA 



and in the third year of his push into the Fields he was able 
to crown a new ambition by the purchase of a block of four 
claims in one of the best-paying sections of the Kimberley mine. 
His savings were then about ^3000, and he put nearly every 
pound he was worth into his purchase. His seemingly risky 
investment was quickly justified by the yield of his claims. 
With the help of this great investment he came swiftly into 
prominence. Entering into partnership with his brother, he 

established the firm of Bar- 
nato Brothers in 1880, as a 
London and Kimberley firm 
of diamond dealers and brok- 
ers in mining properties, and 
crowned a further ambition 
by combining his own claims 
with adjoining holdings in his 
first mining stock organiza- 
tion, " The Barnato Diamond 
Mining Company." 

He was one of many 
quick-sighted and resourceful 
men who perceived that the 
day for any profitable work- 
ing of individual claims had 
passed, while the body of 
miners was still struggling 
along blindly in the great cav- 
ing chasms. He brought about a highly desirable amalgamation 
of the claims which he controlled with those of the Standard 
Company, one of the strongest organizations in the Kimberley 
Mines, and later these claims were amalgamated with the hold- 
ings of the Kimberley Central Company, in which he became a 
large shareholder. It was at this stage in his fortunes that he 
came into keen rivalry with the only competitor that could make 
headway successfully against him, Cecil John Rhodes. 

There was a singular likeness in some respects in the careers, 




C. J. Rhodes, when a Student at Oxford. 



THE MOVING MEN 



2 73 



conceptions, and calculations of these extraordinary men, although 
they were so markedly dissimilar in personal appearance and 
temperament. Cecil John Rhodes was the younger son of a 
Hertfordshire clergyman, and came as a sickly boy to South 
Africa in 1871, in the first flush of the diamond fever, to join 
his brother Herbert on a small plantation in Natal. The raw, 
dusty Diamond Fields were apparently one of the spots least 
likely to attract a youth whose health had broken down, and 
whose tastes were bent from early childhood toward a scholar's 




Cape Town. 

life in the cloisters of a university appealing to every high imagi- 
nation in its memorials of every age since the dawn of letters 
in Britain. So indeed it seemed when young Rhodes turned 
his back on the fresh glitter of the new mines and entered his 
name on the rolls of Oriel College in 1872. But the same year 
saw his return, because of a lung fever that threatened his life, 
and made the shift from misty England to the mild clear air of 
the terraces of Natal an imperative prescription. Shortly after 
his return Herbert Rhodes slid into the current setting to the 
Diamond Fields, but Cecil stayed on the plantation until the 
following year, 1873, when his brother's report and his dawning 



274 THE DIAMOND MINES OF SOUTH AFRICA 

success as a claim-owner drew him, somewhat reluctantly, over 
the long sun-baked stretch to Kimberley. 

So unknown to each other and blind to their future clash 
and union, Cecil John Rhodes, the clergyman's son, and Barney 
Barnato, the London shopboy, started abreast in the race for 
fortune on the same track. An ordinary observer of the two 
young men would probably have picked Barnato as the winner 
on such a track as the new Diamond Fields. Any one could see 
at a glance that the young Hebrew was unsinkable, and pecul- 
iarly fitted to make a good living in the stirring towns by his 
business training, quick wit, and racial genius for trade, while 
the English college student had no apparent fitting for success 
either as a digger or a business man. Kipling has told of the 
straining of the new ship, as a living thing, in the trial to find 
herself, and this fine conception has literal truth in the applica- 
tion to young manhood. So Cecil John Rhodes was forced to 
find himself, as he did, when he put away his books to plunge 
into the whirling life of the Great White Camps. 

Tall, gaunt, shy, the stripling sat at the diamond sorting 
table, overseeing the Kafirs who scraped over the pebbles from 
his brother's claim, on a little " floor " near the edge of the big 
Kimberley pit. Roughly dressed, coated with dust, disdainful 
of any foppish touch, peculiarly self-contained, full of novel ideas 
and aspirations rising, turning, and shaping themselves in his 
mind, he was not one to mingle, like Barnato, in every stir 
of the froth in the camps, or ready to jump, like the London 
shopboy, into any gush of speculation, from a bet at cards to an 
auction sale. Externally the two young men could scarcely be 
more unlike than the little, chunky, bullet-headed, near-sighted, 
mercurial Hebrew, taking a hand in current sport or traffic, and 
the tall, thoughtful, young overseer, sitting moodily on a bucket, 
deaf to the chatter and rattle about him, and fixing his blue eyes 
intently on his work, or on some fabric of his brain. 

Yet both were alike in their expanding ambition and power 
to grapple and mould in their distinctive ways the opportunities 
about them. Both had keen foresight, and extraordinary com- 



THE MOVING MEN 



275 




Silver Trees. (These trees grow only on the slopes of Table Mountain.) 

prehension of great financial undertakings. Both had, too, the 
essential poise and accuracy of judgment that shuns pitfalls 
and punctures illusions. With variant motives they sought the 
same end of great riches : one for the sheer satisfaction of money 
making, of unfolding great schemes of production and flotation, 
of proving to the world that he was a master of finance ; the 
other chiefly as a means to reach ends of Imperial scope, to 
throw the searchlights of civilization into every cranny of the 
Dark Continent, to lift the prodigious dead weight of unnumbered 



276 THE DIAMOND MINES OF SOUTH AFRICA 

bygone ages of barbarism, to make the waste lands fruitful and 
open the arteries of traffic, to create a Greater Britain than the 
most daring fancy before him had conceived, and stretch the 
hand of his Queen over a realm transcending the farthest sweep 
of the Macedonian or the Roman. 

Both realized very keenly the practical necessity of effecting 
combinations of the claims covering the diamond mines in order 
to provide a uniform and efficient development and to secure 
a scarcely less essential control of the diamond output. The 
patent collapse of the open pit mining forced the undertaking 
of underground works, and compelled the further consolidation 

of holdings ; but 
for too many years 
there was no com- 
mon realization of 
the urgent need of 
the systematic de- 
velopment of the 
mines as a united 
property, and not 
as a complex col- 
lection of discord- 
A Cape Can. an t parts. The 

working of the parts was at best cramped and conflicting. The 
prosecution of any well-designed plan was heavily handicapped 
by the lack of cooperation in adjoining properties. 

This was sharply etched in by Barnato after Rhodes had 
successfully pressed the amalgamation of the variant interests. 
" I think I can prove to you, gentlemen," he said, in addressing 
a shareholders' meeting in 1889, "that in order to work the 
underground system, you must have the mines intact. You all 
remember the trouble and friction that took place when the De 
Beers mine was being worked by the De Beers Company, the 
Victoria, the Oriental, the Elma, the Gem, and others. Why 
was the underground system not a success in this case? Because 
one company was working against another ; that is to say, if one 




THE MOVING MEN 



277 



company was on the five hundred feet level, the opposing com- 
panies could go and eat into each other's boundary walls and 
pillars to such a dangerous extent that the entire mine was in 
a condition which threatened collapse at any moment." 

This was so patently true, and more particularly in Kimber- 
ley mine, that it may seem surprising that the disastrous conflict 
was so long maintained. But it must be borne in mind that 
the average shareholder was not as quick to see and prompt to 
move for a remedy as Rhodes, and comparatively few had his 
intimate and comprehensive knowledge of the condition of all 
the mines in the Fields. A very large proportion of the investors 
in these mines were men who had never been on the Fields at all, 
or whose acquaintance was limited to a sightseer's visit. Many, 
too, had bought shares simply as a gamble in the stock market, 
and only welcomed such information or reports as were calculated 
to boom their speculations. 

It was obviously labor lost to attempt to interest such men 
in any far-reaching plan for the union and systematic develop- 
ment of all the mining claims in the craters, and most of them 
would have sneered it away as a mere chimera if it had been laid 
before them. This was indeed a project which might well have 
appalled an ordinary man, even if he had the clear sight and 
comprehension of the position essential to a true judgment. 
Anybody might dream of such a gigantic combination, and some 
day-dreamer might babble about it to his gossips, but what man, 
or association of men, would have the foresight and patience, the 
perseverance and tact, the integrity and fulness of talent, to push 
forward toward it for years, to thrust aside or crush blocks in 
the way, to harmonize discordant and jealous interests, to open 
the eyes of narrow-sighted selfishness, to win the confidence of 
the distrustful, to design a scheme of union that would make all 
holders of good working claims common shareholders on a basis 
of equity and assured profit to all, and finally to provide the 
enormous capital necessary for the consummation of the scheme, 
and the development of the great diamond mines in a really 
great way ? 



278 THE DIAMOND MINES OF SOUTH AFRICA 

Here was a task of such tremendous magnitude and difficulty 
that men of good ordinary judgment might well question its 
feasibility. What man in or out of the Fields would dare attempt 
it ? Who could do it, if he dared to venture ? There is a mighty 
fillip to the conceit of man, that in such great exigencies as these 
in times when some prodigious undertaking is imperatively 
needed the man or men who can carry it on to completion 
are almost always forthcoming. " Nothing is impossible nowa- 
days," said the " Bonanza King," Flood, when doubts were 
raised of the practicability of piping water from the Sierra 
Nevada Mountains to the Comstock Silver Mines on the 
Virginia range ; " the only question is, will it pay ? " That 
seems, indeed, the only touchstone which men of such pith and 
temper are disposed to apply to any object. It was again made 
evident on South African Diamond Fields how far the possible 
stretches when men with Flood's touchstone are the adven- 
turers. The moving men, who could comprehend the need for 
union and effect it, came irresistibly to the front in the Fields. 

The undertaking to which they set their hands should be 
clearly set forth. In spite of the ruin of the open mine work- 
ings in the competing development scramble, and in spite of the 
continuing conflict and recurrent disasters in the underground 
mining so cogently enforcing the call for union, there were 
still, at the end of 1885, no less than ninety-eight separate hold- 
ings in the four mines. In Kimberley mine there were eleven 
companies and eight private holdings; in De Beers there were 
seven companies and three private holdings ; in Dutoitspan, six- 
teen companies and twenty-one private holdings ; in Bultfontein, 
eight companies and twenty-four private holdings. Thus the four 
mines were operated by a total of forty-two companies and fifty- 
six private firms or persons, all clashing within a surface area 
of 70 acres. The original location claims, aggregating 3600, 
had been united to this extent, merely, at the close of fourteen 
years of mining on the helter-skelter plan. 

It is hardly just to credit Rhodes and Barney Barnato with 
an equal perception of the imperative call for the union of all 



THE MOVING MEN 



279 




Mr. C. D. Rudd. 



the discordant interests in the diamond mines. Each reached 

the conclusion that it was no longer possible to continue to 

work the mines divided into small 

holdings which were controlled by 

men antagonistic to one another. 

Rhodes's interests were mostly in 

De Beers mine, and Barnato's 

largely in the Kimberley mine. In 

the same year, 1880, in which Bar- 

nato floated successfully his first 

diamond-mining corporation, "The 

Barnato Mining Company," con- 
sisting of a few claims in a rich 

section of the Kimberley mine, 

Rhodes and others founded the De Beers Mining Company, 

on the contiguous diamond-bearing crater. 

It is of interest in this connection to trace the origin of De 

Beers Mining Company through 
the early years of De Beers mine. 
In 1 873 Rhodes united his claims 
in De Beers mine with those of 
C. D. Rudd, and they slowly in- 
creased their holdings. Robert 
Graham joined them in 1874, 
and later Runchman, Hoskyns 
& Puzey took part with them in 
the purchase of Baxter's holdings. 
This combination, in addition to 
mining their own ground, took 
pumping contracts to drain the 
mine. Besides the above combi- 
nation there were other competi- 
tors for the purchase of claims, 
such as Dunsmure & Alderson, 

Stow & English, and these three firms gradually acquired all the 

best ground in De Beers mine except the Elma Company, owned 




Mr. Robert English. 









28o THE DIAMOND MINES OF SOUTH AFRICA 

by Thomas Shiels and others, the Victoria Company in which 
J. Ferguson was then the leading spirit, and the United Dia- 
mond Mining Company. 

The De Beers Mining Company was formed on the ist of 
April, 1880, with a capital of ,200,000, by the union of the 
three firms first mentioned. It progressed with extraordinary 
success, extending its range of ownership, absorbing step by step 
its floundering neighbors, and finally standing out preeminent 
in March, 1885, with a capital of ^841,550, upon which divi- 
dends of ~i\ per cent had been paid during the last fiscal year, 
in spite of the heavy charges of development work and the un- 
avoidable hampering of its mining operations. Mr. Rudd states 
that at one time Rhodes and he had the offer of the entire De 
Beers mine for ^6,000, and they walked about a whole day 
talking it over, but finally decided they could not finance it. 
The licenses at that time were so costly that it was thought wise 
not to risk the purchase. Money was not very plentiful among 
these men in those days, as is shown by one of the first checks 
of the De Beers Mining Company, which was drawn by Rhodes 
in his own favor for ^5, "as an advance against his salary as 
secretary." 

It is possible that Barnato may have tried to bring about a 
further consolidation of some of the various interests in Kim- 
berley mine, but there is nothing to show that he contemplated 
any broad scheme. 

For nearly six years Rhodes concentrated his efforts in the 
Diamond Fields toward obtaining complete control of De Beers 
mine by himself and his chosen friends, and he brought about 
this consolidation of all the holdings in May, 1887. His 
master mind was steadfastly bent on the attainment of the con- 
trol of the development and output of the four great diamond- 
producing mines of South Africa, and his work of first uniting 
all the interests in De Beers mine was but the beginning of his 
great dream. The range for amalgamation of the four mines 
was so great that no single man, however ambitious, could hope 
to cover it by any single-handed effort. The consolidation of 



THE MOVING MEN 281 

all the companies in De Beers mine was on the lines conceived 
by Rhodes, and carried out by the support given him by the 
leading men who were interested in the various companies. 

Up to this time there was no rivalry between Rhodes and 
Barnato, for no measures had been taken by Rhodes to obtain 
a footing in Kimberley mine. The first steps taken in this 
direction were to try to purchase the claims in the west end of 
the Kimberley mine held by the Cape of Good Hope bank, 
and known as W. A. Hall's claims. This was in the beginning 
of May, 1887. Unfortunately, however, for Rhodes's scheme, 
these claims had already been offered to a syndicate in London, 
headed by Sir Donald Currie, and were purchased by that syndi- 
cate for i 10,000. The plan which Rhodes had in his mind 
was to purchase these claims, and also to purchase the claims 
of the " Compagnie Fran9aise des Mines de Diamant du Cap 
de Bon Esperance," known as the " French Company." The 
"French Company" held a block of claims which ran nearly 
across the mine from north to south, and divided the holdings 
of the Central Company. It also held a block of claims adjoin- 
ing those of W. A. Hall, but these were not connected with the 
main body of their claims, being separated by the intervening 
claims of the Central Company. These two companies were 
so antagonistic to one another that neither would allow the 
divided blocks of ground to be worked by means of tunnels 
driven through the diamond-bearing ground of the opposing 
company. The Central Company worked its claims by two 
separate shafts sunk in the blue ground at the bottom of the 
open mine, and the ground hoisted in the shafts was sent to the 
surface by means of aerial trams, while the " French Company " 
was compelled to drive tunnels into the walls of the mine adjoin- 
ing the claims and connect them by a cross tunnel, as they were 
working through one shaft only. 

To create a powerful company in Kimberley mine was sub- 
stantially all that the leading men in that mine had been work- 
ing for, but this was far from satisfying Rhodes. Barnato viewed 
the situation as a speculator and investor. Money making 



282 THE DIAMOND MINES OF SOUTH AFRICA 



M 

_ : v ^L_iil ^y f 




THE MOVING MEN 283 

through mining on a sound basis was avowedly the limit of his 
scheme, apart from a natural pride in figuring as the foremost 
operator in these marvellous Diamond Fields, and a rising star 
of the first magnitude on the London Stock Exchange. But the 
assurance of money making was, at most, a minor consideration 
with Rhodes. He, too, valued money highly, but not for the 
bare delight in piling it up or for the luxuries which it would 
purchase. Great wealth was to him the essential means for the 
furtherance of great plans. He wanted millions in hand, or the 
assured control of millions, to push his design for the lighting-up 
of the Dark Continent by the torchbearers of civilization, for the 
carrying of the flag of Greater Britain from the Cape to Cairo. 

A man of kindred spirit, but of far more quixotic temper, 
the great soldier, General Gordon, once told him of the offer of 
a roomful of gold by the Chinese Government for his extraordi- 
nary services in subduing the Tai-Ping rebellion. 

" What did you do ? " said Rhodes. 

" Refused it, of course," said the disdainful Gordon. " What 
would you have done ? " 

" Done," said Rhodes, " why, I would have taken it, and as 
many more roomfuls as the Chinese would give me. It is no 
use to us to have big ideas, if we have not got the money to 
carry them out." 

The range of his plans and how he pursued them will be 
presented in detail in the chapter dealing with the far-reaching 
undertakings of the great Chartered Company which he con- 
ceived and brought into existence. It is sufficient to note at 
present that he pushed the development of his grand political 
aims apace with the means at his command, from the very begin- 
ning of his appearance as a prominent factor in the development 
of the diamond mines. He entered the Cape Parliament as 
a member for the district of Barkly West, almost coincidently 
with the formation of the De Beers Mining Company. From 
the day of his entrance into the political field, he worked un- 
waveringly for the extension of British dominion into the heart 
of Africa. 



284 THE DIAMOND MINES OF SOUTH AFRICA 




THE MOVING MEN 



285 



The northern boundary of the province of Griqualand West, 
formed by the inclusion of the new Diamond Fields, had not 
been determined by careful surveying, and the location of the 
line was disputed by the Batlapin chief, Manlcoroane, who 
claimed control of the territory which is now Lower Bechuana- 
land. Rhodes prevailed on the Cape Government to form 
and send out a Delimitation Commission for the settlement of 




Avenue of Oaks, Cape Town. House of Parliament at the Left. 

the dispute, and his appointment as one of the commissioners 
was a natural recognition of his interest and competence. 
Shortly after he reached the frontier he was able to satisfy 
himself that the complaint of the chief was well founded. Some 
seventy farms belonging to Mankoroane's tribe had been in- 
cluded in error within the bounds of the British province, and 
justice demanded this acknowledgment. But instead of aban- 
doning the ground, Rhodes saw that restitution might be 
made in a way to accord with his aim for the extension of 



286 THE DIAMOND MINES OF SOUTH AFRICA 

British sovereignty, and his cogent appeal persuaded the 
Batlapin chief to place all his territorial holdings, covering half 
Bechuanaland, under British protection by cession to the Cape 
Colony. To his mortification, however, the Colony declined 
the offered cession with its contingent obligations. Then 
Rhodes appealed to the Home Government, and finally suc- 
ceeded in obtaining the establishment of a Protectorate over 
Lower Bechuanaland in 1884. 

But it was only by the most pressing insistence that this 
advance was maintained. The Cape Colony was so stubborn in 
its refusal to bear the expense of any new acquisition, and the 
Imperial Government was so doubtful and sluggish in grasping 
its opportunities, that Rhodes was forced to the conclusion that 
the only assurance of the accomplishment of his aims must 
come from his own private enterprise, through the forming of 
some great corporation, whose capital and interests might be 
engaged in his undertaking for the control and development of 
the resources of the vast barbaric interior of Africa. It was 
for this cause chiefly that he was so unflaggingly insistent upon 
the farthest possible stretch of amalgamation in the control of 
the diamond mines, though it must justly be observed that the 
thorough amalgamation of conflicting interests in the Fields was 
very highly desirable, if not absolutely essential, for its systematic 
development and the marketing of its output. A possible com- 
bination, with which Barnato would have rested content, would 
have wholly failed to accomplish the end which Rhodes had so 
deeply at heart. 

In the year 1887, shortly after taking charge of the De 
Beers Mining Company, Mr. Rhodes requested me to write to 
two of my friends in London, Mr. Hamilton Smith and Mr. 
E. G. De Crano, who founded the Exploration Company of 
London, and who were intimately connected with the Messrs. 
Rothschild, and request them to ask Lord Rothschild if he 
would supply the funds for the purchase of the French Com- 
pany in the Kimberley mine, provided Rhodes could come 
to some agreement with that Company for the purchase of the 



THE MOVING MEN 



287 



property. Before any answer could be received, even by cable, 
Rhodes, who had gone from Kimberley to Cape Town to 
attend the Session of Parliament, became very impatient about 
securing this property, and wired me to join him, and we sailed 
from Cape Town on the 6th of July. In my letter to Messrs. 
Smith and De Crano I put before them the plan which Rhodes 
proposed to carry out, and the object he had in purchasing the 
French Company's property, viz., to prevent the amalgamation 
of all the interests in that mine, which might be set up as an 
independent company in conflict 
with the interests of De Beers. 

On our arrival in London we 
met Lord Rothschild, and Rhodes 
discussed the plan with him. In 
the meantime, while we were on 
the water, Rhodes's scheme had 
been presented to the late Mr. 
Tite and to Mr. Carl Meyer of 
Messrs. N. M. Rothschild & 
Sons, who were very favorably im- 
pressed with the business, and had 
discussed it with Lord Rothschild. 

Mr. De Crano had made several '. 

trips to Paris, and had already Mr. carl Meyer. 

paved the way for a conference between Rhodes and the directors 

of the " French Company." 

At the close of the interview, Lord Rothschild said, " Well, 
Mr. Rhodes, you go to Paris and see what you can do in refer- 
ence to the purchase of the French Company's property, and in 
the meantime I will see if I can raise the _ 1,000,000 which you 
desire." 

On leaving the room Lord Rothschild stopped Mr. De 
Crano for a moment, and said to him, "You may tell Mr. 
Rhodes that if he can buy the French Company, I think I can 
raise the million pounds sterling." 

The same evening Rhodes, Mr. De Crano, Mr. Harry 




288 THE DIAMOND MINES OF SOUTH AFRICA 

Mosenthal, and myself left for Paris, and after several meetings 
with the French Company's directors, we settled upon the terms 
for the purchase of their property, which they valued at that time 
at about _ 1,400,000, including all their assets. On returning 
to London Mr. Rhodes arranged with Lord Rothschild that he 
should furnish him with ^750,000, which would be sufficient for 
the time being to complete the arrangements that he had made 
with the French Company. In my letter of the i8th of June, 
it was mentioned to Messrs. Smith and De Crano that Rhodes 
would be willing to issue De Beers shares in payment of the 
loan at ^i less than the ruling market price of the shares at the 
date the money was paid, and would pay Messrs. Rothschild a 
handsome commission for transacting the business. 

The final arrangement made for the payment of this money 
was the issue of 50,000 De Beers Mining Company's shares at 
^15 per share, and a syndicate was formed to take up these 
shares with the able assistance of Mr. Ludwig Lippert, of Ham- 
burg. It was agreed between Lord Rothschild and Rhodes 
that the profit on the rise of the shares between ^16 and .20 
during the next three months should be divided between the 
purchasing syndicate and the De Beers Company. The shares 
rapidly rose, and, before the expiration of the time, had already 
reached ^22 per share. The De Beers Company received 
^100,000 as their portion of the profit on the rise of the shares. 
Shortly after the completion of this business Rhodes returned 
to the colony and awaited the result of the French Company's 
shareholders' meeting to confirm the sale which had been made 
to him by the directors of that company. Barnato and others 
interested in the Kimberley Central Company, upon hearing of 
the transaction that had taken place, determined to use every 
effort to prevent the consummation of this sale, and threatened 
to offer the shareholders of the French Company at their gen- 
eral meeting ^300,000 more than the amount for which the 
directors had pledged the company to Rhodes. 

As a general of a great army is obliged to have the assistance 
and cooperation of competent lieutenants to carry out the plan 



THE MOVING MEN 



289 



of campaign which his superior mind has conceived, so Rhodes 
looked about for the strongest and ablest men to join him in 
repelling the vigorous attack which was being made against him. 
The first check which he gave his opponents seemed at first sight 
to be a complete surrender to them. Instead of allowing Bar- 
nato and his colleagues to bid against him for the purchase of 
the French Company, Rhodes arranged with them that he should 
complete the purchase upon the lines agreed upon with the direc- 
tors of that company, and promised to unite the interests so pur- 
chased with the Kimberley Central Company, in which Mr. Francis 
Baring-Gould, who was the chair- 
man, Barnato, and others held the 
controlling power, taking shares in 
the Central Company in payment. 

In this, as well as in subsequent 
transactions, Rhodes was most ably 
assisted by Mr. Alfred Beit, the 
Kimberley representative of Jules 
Forges & Co., who started business 
in Paris as diamond merchants in 
1869. The men who from time to 
time have been connected with Mr. 
Forges and the successors to him, 
Messrs. Wernher, Beit & Co., took 
the keenest interest in Rhodes's scheme, and assisted him more 
than all others in bringing about the consolidation of the dia- 
mond interests. As early as 1871 Mr. Julius Wernher went 
out to Kimberley in the capacity of diamond buyer for Jules 
Forges & Co., and became partner in the firm in 1878. The 
firm grew in importance, and became owners in some of the 
largest companies in the four mines. They were the founders 
of the Griqualand West Diamond Mining Company in Kim- 
berley mine, which was afterward re-formed into the " French 
Company." Mr. Alfred Beit came to the fields in 1875 as a 
diamond buyer for the firm of Lippert & Co., of Hamburg, and 
after a few years established himself in business as a diamond 




Mr. Alfred Beit, while a Resident of 
Kimberley. 



290 THE DIAMOND MINES OF SOUTH AFRICA 

buyer on his own account. In the year 1882 he joined the firm 
of Jules Forges & Co., as their representative in South Africa, 
and became a partner in the firm in 1886. 

In 1889 Mr. Forges retired from the firm, which was re- 
formed as Wernher, Beit & Co., Mr. Max Michaelis joining 
the firm. Mr. Michaelis came to the Fields in 1878, and went 
into partnership with Mr. S. Neumann. He organized the Cape 
Diamond Mining Company in Kimberley mine. In 1880 he 
entered into an arrangement with Jules Forges & Co. to carry on 
his diamond business on joint account with them, which arrange- 
ment remained in force until Mr. Forges retired, when he became 
a partner in the new firm. Mr. Michaelis assisted in bringing 
about the fusion of several of the large claim-holders in the 
Kimberley mine, such as Baring-Gould & Atkins, and Baring- 
Gould, Price & Tracy, with the Kimberley Central Company. 

The great initiative and business capabilities of Mr. Beit 
were heartily recognized by Rhodes, and he was very largely 
instrumental in building up the diamond-mining industries, and 
bringing the dreams of Rhodes into practical shape and on 
business lines. 

At a special general meeting of the shareholders of the De 
Beers Mining Company Limited, held at Kimberley on the jist 



of March, 1888, for the purpose of considering 
firming an agreement entered into between the 



and con- 
respective 




The Diamond Market, Kimberley, 1875. (First Office of Mr. Alfred Beit at the Left.) 



THE MOVING MEN 291 

Boards for the amalgamation of the De Beers Mining Company 
with De Beers Consolidated Mines Limited, Mr. Rhodes gave 
his reasons for the necessity of acquiring either the control of the 
Kimberley mine or of entering into some arrangement with the 
directors of the Central Company, who controlled the mine, by 
which the output of both De Beers and Kimberley mines could 
be regulated. He saw that by skilful and systematic mining 
on the underground system, the output of the mines could be 
increased far beyond the world's requirements. It was clear, 
too, if these two mines were run in opposition to one another, 
it would result in the flooding of the market with diamonds, and 
a consequent depreciation of their value, with a fall in market 
prices almost ruinous to both companies. He saw that the out- 
put of diamond-bearing ground could be made almost unlimited, 
and in referring to this he said: "We had to face either an arrange- 
ment with the Kimberley Central Company, or obtain control 
of the Kimberley mine. We approached the Kimberley mine 
management in every possible way we could conceive. I valued 
the De Beers mine higher than they did, but I was willing to 
give way in order to obtain control. I was met simply with 
smiles and obdurate statements. I was met with the arguments 
of the gentleman at ' the corner,' who said the Kimberley mine 
was worth three times as much as De Beers. We had to choose 
between the ruin of the diamond industry or the control of the 
Kimberley mine. We saw this, that you could never deal with 
obstinate people until you got the whip hand of them, and that 
the only thing we had to do to secure the success of our industry 
was to get the control of the Kimberley mine." 

As soon as Rhodes had bought the French Company and 
amalgamated his interests with the Kimberley Central Company, 
he found that the management of that Company was headstrong 
in its determination to run the Kimberley mine in rivalry with 
De Beers. This was diametrically opposed to his conviction 
that monopoly was the essence of success in diamond mining ; 
for, as he said, " Our engineers had long ago shown us that, by 
underground working, Kimberley and De Beers mines could 



292 THE DIAMOND MINES OF SOUTH AFRICA 




THE MOVING MEN 293 

produce far more diamonds than the world would take." By 
the purchase of the French Company, De Beers Company held 
one-fifth of the capital of the Central Company, and after many 
attempts to bring about a friendly union of the two mines, 
Rhodes came to the conclusion that the only feasible plan 
was to buy a sufficient number of shares in the Central Company 
to obtain control. To accomplish this would take at least 
^2,000,000 sterling. Fortunately Mr. Alfred Beit, whose com- 
mand of capital for such great undertakings was unequalled in 
South Africa, stood fast by him in determined cooperation. In 
answer to Rhodes's natural question, " Where is the money to 
come from ? " Mr. Beit said pithily, " We will get the money 
if we can only get the shares." 

Then ensued a most keen contest. Mr. Beit and Rhodes be- 
gan buying all Central shares that could be secured with apparently 
limitless means. Both were leaders in the contest, but Mr. Beit 
furnished most of the money. Meanwhile Barney Barnato was 
bidding against them with unfailing pluck for the control of the 
Company. The price of shares mounted by jumps, but never 
too high for Barnato, who was persistent in his claim that the 
Kimberley mine was worth two of De Beers. Rhodes's version 
of the story of this struggle is that in his purchase of shares he 
had the support of the loyal directors and shareholders of his 
Company, while his principal opponent was handicapped by the 
fact that he was forced to buy out his own largest shareholders. 
There is little doubt that Barnato felt this apparent lack of loy- 
alty keenly, but he was too strenuous a fighter to concede defeat. 
As a matter of fact, he came to terms with his antagonists only 
when the price of shares had been bulled to a figure that seemed 
out of reason even to his sanguine estimate, while the price of 
diamonds had been forced down unprofitably by unobstructed 
competition. After many and long conferences, Rhodes made 
Barnato one last offer, which he accepted. For his interest in 
the Kimberley Central Company he was paid with De Beers 
shares at the current rates of shares on the day of the sale. By 
this purchase De Beers' holding of Central shares was brought 



294 THE DIAMOND MINES OF SOUTH AFRICA 

up to eleven thousand out of seventeen thousand shares. Under 
the trust deed of the Central no amalgamation could be made 
unless half the capital was present at a meeting called for the 
purpose, and no new resolution could be carried without a two- 
thirds majority of those present. The bargain with Barnato 
gave De Beers the control. 

So having finally obtained the control of the Kimberley mine 
by purchase for .5,338,650, Rhodes turned his attention to 
what he called the poorer mines, Dutoitspan and Bultfontein. 
At a meeting of De Beers shareholders he said he was reminded 
of a story he had read about a certain mine, of which it was said 
" it was too rich to leave and too poor to pay," and he would 
thus describe the mines alluded to. " Nothing," he said, " was 
so extraordinary as the way in which the people would hold scrip 
from year to year that never pays, but it was always said, ' Oh, 
next year it will pay,' and so it went on from year to year." He 
wished to state " that so far as the amalgamation of the diamond 
mines was concerned, it would not help the poorer mines, but 
rather the other way. It was generally noticed in mining 
matters that following upon one success a number of unsuccess- 
ful ventures were floated. And this was why they had secured 
these interests in Dutoitspan and Bultfontein mines." He did 
not look upon the purchase of properties in these mines as a 
good investment, with diamonds at the price they were bringing 
at the time of the purchase ; but as these two mines were large 
factors in the production of diamonds, their yield, even if mined 
at a loss, would affect in a very large degree the price which 
could be obtained for the product of the richer mines. 
Although Dutoitspan mine could not be worked at a profit at 
the market price of diamonds, and the mine had already begun 
to be troubled with reef falls burying the blue ground below, 
still he considered it necessary to get control of the principal 
companies in this mine. In Bultfontein mine, where the reef 
troubles had already begun, there was still a large portion of the 
mine in process of working, and he described it as being " on 
the margin of cultivation." If the reef remained standing, and 




.--.: 



THE MOVING MEN 



295 



oiuwn MNva 3dOH 0009 




296 THE DIAMOND MINES OF SOUTH AFRICA 

the price of diamonds was fair, the mine could be worked at a 
small profit. 

Rhodes continued the purchase of the properties in both 
these mines until the whole of the two mines came into the 
possession of the corporation organized as De Beers Consoli- 
dated Mines. He showed the shareholders in the various 
companies that the fate of the poorer mines lay in his hands, 
because he could produce twice the amount of diamonds the 
world required from De Beers and Kimberley mines alone. 
Even at the low rate of fourteen shillings a carat, he made it 
clear that the richer mines could pay to the shareholders divi- 
dends which would satisfy them. " The poorer mines, ' on the 
margin of cultivation,' would have to accept our offers, or fight 
us on two grounds, larger output and lower rates." 

In his speech at the annual meeting of the De Beers Min- 
ing Company, held at Kimberley on the i2th day of May, 1888, 
Rhodes bore tribute cordially to the essential cooperation of 
Mr. Beit in his great undertaking. 

In moving a vote of thanks to the chairman, his former 
antagonist, Barnato, briefly referred to the struggle which was 
closed by the purchase of his shares in the Kimberley mine. 
He said " no person knew better than he did the labor Mr. 
Rhodes had to convert him into the De Beers Mining Company." 
He could say that day after day and night after night Mr. 
Rhodes was laboring to get him to take De Beers for Centrals. 
He gave way .when he saw diamonds down to eighteen shillings 
a carat, and on those conditions he joined Mr. Rhodes. It is 
only just to Barnato to note in closing that he was as loyal in 
his later cooperation as he had been persistent in his antagonism. 
It is sad to recall how his brilliant and versatile mind gave way 
under the enormous strain brought upon him by the various 
obligations incurred through his numerous investments and 
flotations in the gold fields. His tragic death was a distressful 
close to his phenomenal career. On his way to England from 
the Cape, in June, 1897, he suddenly sprang overboard and was 
drowned. 



CHAPTER X 




THE ESSENTIAL COMBINATION 

T has been told why and how the conflicting 
interests on the Diamond Field were fused in 
one dominant organization. The signal ser- 
vices of this amalgamation are now too obvious 
for dispute. By the formation of De Beers 
Consolidated Mines Limited, it became practi- 
cable to design and conduct mining operations systematically 
and economically and to regulate the output to the market de- 
mand. It was soon apparent, too, that the organization of this 
extraordinary joint stock company was the creation of a power 
of yet unmeasured service for the development of the resources 
of South Africa and the push of civilization through the Dark 
Continent. 

The only approaches to the far-reaching conception of this 
organization must be traced back to the old Dutch and English 
East India companies, or the visionary project of John Law, 
exploding in air as the Mississippi Bubble. At the outset, on 
the 1 2th of March, 1888, a seemingly unpretentious joint stock 
company was formed and established at Kimberley with a capital 
of _ 1 00,000 sterling, divided into twenty thousand shares of 
^5 each. Authority was granted, however, in the articles of the 
association, to the shareholders of the company to increase this 
small capital in general meeting, from time to time, for the 
acquisition of new property, by creating new shares to any 
extent, or, in the exact words of article 39, "such amount as 
may be deemed expedient." No provision for expansion and 
acquisition could be more liberal, and the particular specifica- 
tions of the articles of the association show that " new property," 

297 



298 THE DIAMOND MINES OF SOUTH AFRICA 

in possible range at least, was not confinable to the Diamond 
Fields or Cape Colony, or even, perhaps, the scope of the whole 
Dark Continent. 

It was remarked somewhat caustically at the time, but with 
undeniable keenness, that it was much easier to tell what this 
amazing Company could do than to determine what it should 
not do under its articles of association and trust deed incorpo- 
ration under the limited liability laws of the Cape of Good 
Hope. It might shift its head office from Kimberley to any 
other place on earth. It might " acquire by purchase, amalga- 
mation, grant, concession, lease, license, barter, or otherwise any 
houses, lands, farms, tracts of country, quarries, mines, mining 
or other claims, rights and privileges, water rights, waterworks 
or other works, privileges, rights and hereditaments, diamonds 
and other precious stones, gold and other minerals, ores, coals, 
earth, and any other valuable product or substance, machinery, 
plant, utensils, trade marks, patents for invention, licenses to 
use any patented invention, and other movable and immovable 
property of any description in Africa or elsewhere." Under 
this liberal license, the only apparent obstacle to its ownership 
of the face of the earth is the declination of other holders to sell 
or give it away. 

It was further specifically authorized to carry on a mining and 
general trading business in any part of the globe, and to con- 
struct, maintain, and operate any tramways, railways, roads, tun- 
nels, waterworks, canals, gas works, electric works, reservoirs, 
water-courses, furnaces, stamping works, smelting works, fac- 
tories, and in general, "any other works and conveniences which 
the Company may think conducive to any of its objects." It 
might also become interested in, promote, and undertake the 
formation and establishment of such institutions or companies 
(trading, manufacturing, banking, or other) as may be considered 
to be conducive to the profit and interest of the Company, and 
to carry on any business, in short, "calculated directly or indi- 
rectly to render any of the Company's properties or rights for 
the time being profitable." There was also provision for the 



THE ESSENTIAL COMBINATION 299 

possible acquirement of any tract or tracts of country of any size 
in Africa or elsewhere, together with any rights that might be 
granted by the rulers or owners thereof, and the expenditure of 
any sums deemed requisite and advisable in the development 
and maintenance of order and good government in such acquisi- 
tions. 

In view of the enjoyment by the shareholders of such privi- 
leges and liberties, it was only natural that the directors of the 
Company should not be grudgingly confined. This was, indeed, 
the case, and two specifications of powers, in particular, have 
proved to be highly serviceable in practice, for there has been 
no abuse of discretion. The directors were authorized " to pur- 
chase, hire, or otherwise acquire for the Company any share in 
any kind of joint stock company, property rights, or privileges 
which the Company is authorized to acquire, at such price and 
generally on such terms and conditions as they may think fit; 
also to sell, lease, abandon, or otherwise deal with any shares, 
property rights, or privileges to which the Company may be 
entitled, on such terms and conditions as they may see fit, and to 
amalgamate with any other company or companies having objects 
altogether or in part similar to the objects of this Company." 
They were further empowered " to found, promote, float, and 
acquire interest or shares in any companies, undertakings, or in- 
stitutions, as they may deem advisable in the interests of the 
Company ; also to acquire interests in, promote, aid, or subsidize 
any useful industry or undertaking in any country where the 
Company may be carrying on business." 

At the outset, De Beers Consolidated Mines Limited was 
preeminently what is termed a close corporation. Four men 
held all but twenty-five shares of its stock. These, in the order 
of their subscription to the articles of association, were Alfred 
Beit, holding 4439 shares ; Barnett I. Barnato, holding 6658 ; 
Cecil J. Rhodes, holding 4439 ; and Frederick S. P. Stow, hold- 
ing the same number as Beit and Rhodes. These four, by the 
articles, were practically authorized as shareholders to create " five 
life governors or permanent directors of the Company, four 



300 



THE DIAMOND MINES OF SOUTH AFRICA 



of whom shall be Cecil John Rhodes, Barnett Isaac Barnato, 
Frederick Samuel Philipson Stow, and Alfred Beit." If "so 
minded," these four had the power by unanimous resolution of 
themselves or their survivors to appoint the fifth authorized 
" life governor," and to fill any vacancy occurring in their 
number by reason of death or otherwise. These four were 
further constituted the first directors of the Company, and had 




A Group of Directors, De Beers Consolidated Mines Lt'd. (Mr. Frederick Samuel Philipson 
Stow in the centre, holding a book.) 

power to appoint other directors, if they so desired, to act in 
conjunction with them until the first ordinary general meeting 
of the Company, when the shareholders were called upon to de- 
termine how many directors there should be besides the life 
governors, and to elect "such number as they determine to be 
necessary." 

From the point of view of ordinary investors in ordinary 
stock companies the unlimited sweep of this unique organization 
and the powers confided to its controlling directors may be 



THE ESSENTIAL COMBINATION 301 

summed up in the familiar outcry of Dominie Sampson. They 
are indeed " prodigious," but the phenomenal success of this 
combination is a stubborn fact that must be faced in any conten- 
tion that its scope and method of conduct were unwarrantable 
and unadvisable. Its base of operation was not Lombard Street, 
but the heart of South Africa, in a field so unique, in a situ- 
ation so perplexing, in unavoidable touch with such far-rang- 
ing and conflicting interests, that ordinary limitations, hampering 
freedom of expansion and action, would have been crippling and 
possibly disastrous handicaps. The powers of the directors are 
great, but who can justly deny that they have been greatly used 
for the reconciliation of jarring interests, the comprehensive and 
rational development of the diamond mines, the safety and com- 
fort of the miners, the profit of the shareholders, the promotion 
of allied industries, and the general welfare of South Africa ? 
The possible range of expansion of the interests of the corpora- 
tion is a bugbear to some good people, who would prefer the 
harmlessness of the deserted village to the risk that civilization 
might " git forrid sometimes upon a powder cart." But what is 
there to show, to-day, of the actual stretch and exercise of the 
corporate powers beyond the judicious limits of profitable in- 
vestment, sagacious development of tributary resources, and dis- 
charge of patriotic obligations ? 

The expansion of the original corporate foundation was 
rapidly pushed. The plan in detail was presented by the chair- 
man of the corporation, Mr. Rhodes, on March 31, 1888, at 
the special general meeting of the shareholders of the De Beers 
Mining Company. The programme thus presented was unani- 
mously endorsed by the shareholders of the Company, accepting 
it without alteration as the best feasible proposition for the con- 
solidation of the diamond-mining interests. 

At this general meeting of the shareholders the De Beers 
Mining Company was formally merged in the new corporation. 
The shareholders of the old Company received two fully paid 
^5 shares in De Beers Consolidated Mines Limited for every 
fully paid ^10 share in the old Company. Having effected 



302 THE DIAMOND MINES OF SOUTH AFRICA 

this acquisition, transferring the whole of De Beers mine, and the 
interests of the late De Beers Mining Company in all outside 
mining properties, the Consolidated Mines pushed forward stead- 
ily their undertaking of a comprehensive consolidation. The 
first and most important step was the securing of the whole of 
Kimberley mine, the greatest producing factor next to De Beers. 
The method by which the property of the Kimberley Central 
Diamond Mining Company was finally turned over to the Con- 
solidated Mines has been described in the preceding chapter. 

In the acquisition of Dutoitspan and Bultfontein mines a 
different plan was adopted. These mines, at the time, were not 
profitable producing properties, and it was practically certain that 
they could not be operated to advantage in view of the output 
from the greater and richer mines. For several years each of 
these mines had produced diamonds to the value of over half a 
million carats annually ; but this production was rapidly declin- 
ing, owing to the unresisted falls of reef. Among the assets 
taken over from the old De Beers Company were a number of 
shares in the Griqualand West Company of Dutoitspan and 
in the Bultfontein Consolidated Company. By the influence 
secured through this acquisition, it was not difficult to effect 
permanent working agreements with De Beers Consolidated 
Mines, by which the new corporation attained complete posses- 
sion of both mining properties in consideration of the payment 
of a fixed annual dividend. During the second year after the 
incorporation, the Consolidated Mines purchased the property 
of the Anglo- African Mining Company, the Compagnie Generate 
(including its interest in the Conivieras mines in the Brazils), 
the Sultan Diamond Mining Company, and the United Diamond 
Mining Company, representing nearly all the properties of ma- 
terial consequence and extent in Dutoitspan mine except the 
Gordon Company's holdings. During the same period the 
Consolidated Mines bought in the Bultfontein Mining Com- 
pany, the Spes Bona Diamond Mining Company, and the South 
African Diamond Mining Company, comprising a considerable 
part of the properties in the Bultfontein mine. 



THE ESSENTIAL COMBINATION 



303 




304 THE DIAMOND MINES OF SOUTH AFRICA 

The actual cost of the properties thus acquired by the Con- 
solidated Mines was approximately ^14,500,000. There would 
have been no difficulty in expanding the capital of the corpora- 
tion by the issuing of shares to an amount sufficient to cover 
this immense acquisition, but a more conservative course was 
adopted. It was decided not to increase the capital of the cor- 
poration beyond ^3, 950,000. The purchases in excess of this 
issue were provided for by the issue of debentures. The adop- 
tion of this plan necessitated a provision for covering very heavy 
fixed charges in the early years of the operations of the Consoli- 
dated Mines ; but this obligation was undertaken with confidence 
in view of the assurance of the control of the diamond market, 
brought about through the consolidation, and the actual return 
in the rapidly increasing output of the mines with systematic 
and scientific development. 

During the financial year following the completion of con- 
solidation, De Beers produced 2,195,112 carats of diamonds. 
This product, including the proceeds of diamonds from debris 
washing, realized in the market ^3, 287,728. In that year the 
total weight of diamonds produced by all the mines in the 
Kimberley division was 2,415,655 carats. Thus approximately 
ninety per cent of the total production was then furnished by the 
Consolidated Mines. The net profit of the operations for the 
year exceeded _ i ,000,000 sterling, and two half-yearly dividends 
of ten per cent each were paid to the shareholders. The actual 
cost of winning over 2,000,000 carats of diamonds, including 
all expenses at the mines and office charges, was a little over a 
million sterling, or roughly IQJ. per carat. The difference 
between the estimated net profit and the costs of operation 
was expended in the payment of interest on debentures and 
obligations and in provision for their redemption, and in the set- 
ting aside of an exceedingly liberal provision of over ^500,000 
as an offset for depreciation of plant, etc. 

The directors of the De Beers Consolidated Mines could 
point with high satisfaction to this profitable showing in contrast 
with the records of disastrous competition and conflicting mine 



THE ESSENTIAL COMBINATION 305 

operations. No exact statistics are obtainable of the production 
in the early years, when no official returns from the mines were 
made. The late Barney Barnato, who made a special study of 
the probable rate of production, estimated the product from 
1873 to 1880 as ranging annually from a million to a million 
and a half carats. After 1880 there was a considerable increase, 
and in 1883, when official returns were first rendered, the quan- 
tity of diamonds produced was 2,319,234 carats. The average 
value of this product was reckoned at los. ^.\d. giving a total 
of ^2,3 59,466. In 1884 the product was 2,264,786 carats, 
valued at ^2,562,623, showing an average of 23^. i\d. per 
carat. This was the top notch in market value, for in the 
following year, 1885, the diamonds produced amounted to 
2,287,261 carats, with an average value of only igs. $^d. per 
carat. In 1886 the production reached the high total of 
3,047,639!- carats, but the demand increased in proportion, so 
that the average selling price was fully is. higher per carat than 
during the previous year. In 1887 and 1888, through the 
increased facilities for production in De Beers and Kimberley 
mines, the total output rose to 3,646,889 carats, and 3,565,780!- 
carats successively. The average price during these two years 
ranged from iis. 6d. per carat to 12s. \\d. but the market was 
flooded, and prices were falling perilously close to the cost of 
production even in the richer mines. There was no assurance 
of any far-sighted regulation of the output and market prices, 
and, lacking this, diamond mining properties were commonly 
reckoned as little better than gambling ventures. It has been 
clearly shown how this disastrous condition was at once changed 
to stable assurance and prosperity by the control of the new 
organization. 

To the shareholders in the mines, after this reorganization 
was effected, the returns were unprecedented. This profit was 
largely due to the complete control of production, systematic 
operation, and regulation of the output ; but the comparative 
showing was also greatly enhanced by the shrewdness of the 
financiering in the organization, and the withdrawn! of inflation 



306 THE DIAMOND MINES OF SOUTH AFRICA 

from the stocks of the various mining properties included in the 
new incorporation and its leased holdings. The capital of De 
Beers Mines before consolidation was ^2,009,000. The capital 
of the Central Company was .1,779,650. De Beers stock at 
the time of consolidation was selling at 40 a share, represent- 
ing a capital 0^8,036,000. The stock of the Central Com- 
pany, controlling the Kimberley mine, 
was selling at 50 for each 10 share, 
making a total valuation of 8,898,250 
for the mine. At this market estimate 
the valuation of the two great mines was 
17,934,250. The capital of the Du- 
toitspan was approximately 3,500,000, 
and of Bultfontein, 2,000,000 nomi- 
nally, making a gross valuation for the 
four mines of 23,434,250. By consoli- 
dation the capital stock was compressed 
to ^,950,000, and almost absolute con- 
trol of the mining in all four of these great properties was se- 
cured at an annual charge of about ^320,000 for interest on 
debentures and for leases of two companies, one in Dutoitspan 
mine and one in Bultfontein mine. The business of the Com- 
pany grew so rapidly that it was necessary to establish transfer 
as well as general business offices in London. 




Mr. E. R. Tymms, Secretary of 
the London Board, De Beers 
Consolidated Mines Lt'd. 



CHAPTER XI 




SYSTEMATIC MINING 

UST acknowledgment has been made in a former 
chapter of the essential service rendered to the 
diamond mine owners by the device of Mr. 
Edward Jones for underground work beneath 
the fallen reef covering the bottom of the open 
pits. 'This was, however, confessedly only a 
temporary makeshift, enabling the claim-holders to defray the 
heavy costs of sinking shafts through the hard rock outside the 
craters, and pursuing some systematic plan for the extraction of 
the diamond-bearing breccia by underground workings. Deep- 
shaft sinking was undertaken with renewed heart by several 
companies owning claims in Kimberley and De Beers mines, 
but for some years there was an obvious lack of essential 
cooperation and unity of method. Eight shafts were sunk, or 
were under way, in 1885, within and without the craters, for 
opening De Beers and Kimberley mines, and through these 
shafts the blue ground was extracted by four different methods 
of stoping, none of which was satisfactory. The system insti- 
tuted by the Central Company, the largest operator in Kim- 
berley mine, illustrates sufficiently the inherent defects in all. 
Here galleries fifteen feet wide were driven to the right and 
left of a main tunnel, with pillars fifteen feet thick between 
them. Passages or winzes for broken ground were sunk 
at short intervals to a tunnel below. The ground was stoped to 
the height of fifteen feet above the main tunnel, and then below 
it until the stope reached the next level. The passes became 
filled frequently with large pieces of ground, and had to be 
cleared. Under this system the mine was assuming the shape 

37 



308 THE DIAMOND MINES OF SOUTH AFRICA 

of a section of a gigantic honeycomb cut in two longitudinally, 
the spaces for the honey representing the worked-out part of the 
mine, and the comb, the support for the superincumbent mass 
of debris. After a short period of working, the pillars began to 
show signs of crushing, and the mine was considered too danger- 
ous to allow the men to remain in it. They were withdrawn 
just in time to prevent a disaster, for the whole underground 




The Last of Open Mining, Kimberley Mine. 

works collapsed shortly after the last man had left the mine. 
Fortunately no one was killed. The mine had to be reopened 
from top to bottom, for every underground excavation was filled 
up at the close of the year 1888. 

The errors in engineering were further accentuated, during 
the early stages of underground mining, by the jealous bickering 
of rival owners, which was constantly impeding the progress of 
the workings, and it was seemingly impracticable to agree upon 
any plan securing concert of operation and expert opening of the 



SYSTEMATIC MINING 



39 



ROCK SHAFT 



PLAN OF DE BEERS MINE 

7OO FEET LEVEL 
SYSTEM OF WORKING 1887 

THE SQUARES REPRESENT CLAIMS 
81 FT* BY 81 FT. 




ROCK SHAFT 

n 



PLAN OF DE BEERS MINE. 

800 FT. LEVEL 

SYSTEM OF WORKING. 1888 




310 THE DIAMOND MINES OF SOUTH AFRICA 

mines. At the end of the year 1885, although the need of 
amalgamation of claims was obvious and imperative, there were 
still, as has been noted, ninety-eight separate holdings in the 
four mines. Prior to the consolidation of the holdings in De 
Beers and Kimberley mines, the underground workings were 
prosecuted with the general design of withstanding pressure and 
sliding of the reef by leaving sufficient solid blue ground, in the 
form of " floors " or " roofs," between the series of levels, sup- 
ported by buttresses and pillars of blue ground. Costly experi- 
ence by frequent collapses of the roofs and crushing of the pillars 

SECTION THROUGH DE BEERS MINE LOOKING NORTH 



/Spill 

'/' V/MELAPHYRE ' f 

,"//,;(','/,'' /' .7 V"/ 1 *''/' * 




'fizci VC ^1 r~ ^ : ^ iA*ic^':>; 



. . 

100 200 300 400 &QQ 



. .--i. 

'-^ wsfcwio IV IWMIA* * CO., M.Y. 



proved that the levels were too near one another, and that gal- 
leries driven full size from the offsets were difficult to maintain 
and unsafe for the workmen. 

The heavy expense of sinking vertical shafts and driving 
tunnels through the hard rock surrounding the mine had led to 
the adoption of inclined shafts in order to reach the blue ground 
more quickly ; but, for several reasons, these inclines were not 
adapted for the prosecution of deep underground works. The 
chief defects may be briefly summarized. They were difficult 
to maintain, as they were sunk obliquely through the horizontal 
strata of the shale, which frequently gave way and crushed the 
shaft timbers. Secondly, being inclined to the horizon (De 
Beers 56, and Kimberley Standard Shaft 32) and situated not 
far from the margins of the mines, they soon reached blue ground. 



SYSTEMATIC MINING 



H 

o H 

M 

Ul Ul 

Q ^ 



o z 



BHWE 

mmjj 

Milt 
i 



K 




312 THE DIAMOND MINES OF SOUTH AFRICA 

and were continued down in this breccia, which must sooner or 
later be mined. Some of these shafts, as at De Beers, had a 
uniform slope from top to bottom, while others, as at Kimberley 
mine, changed to a steeper slope in depth and in one case to 
a vertical shaft. De Beers No. 2 worked well to the depth of 
800 feet, and the Standard shaft, Kimberley mine, was fairly 
serviceable to the depth of 845 feet. The shafts were not sunk 
with the view of putting in proper pumps, and when steam was 
taken into the mines through them, for pumping purposes, the 



conn 



OUR 



PLAN OF KIMBERLEY MINE 

1000 FT. LEVEL 




natives had to pass up and down the same shafts by means of 
ladders. As all the inclined shafts were upcasts, the heat was 
insufferable. 

When I took charge of De Beers mine, in the year 1887, it 
was worked under what was then known as the Gouldie system, 
which had been copied from the hematite mines of Cumberland, 
and first introduced in the Kimberley mine by Mr. Joseph 
Gouldie, then manager for W. A. Hall, and afterwards mine 
manager of De Beers Mining Company. At De Beers mine an 
inclined shaft had been sunk to the foo-foot level, with inter- 
mediate levels 30 feet apart between the jSo-foot and 5OO-foot 
levels. 



SYSTEMATIC MINING 



SECTION OF KWBERLEY MINE 

LOOKING NORTH 

10060 po 800 SCO 400 600 00 FT. 
6CALE 




SECTION OF KIMBERLEY MINE 
LOOKING EAST 

100 50 100 800 800 400 BOO FT. 
SCALE 




3H THE DIAiMOND MINES OF SOUTH AFRICA 

The plans on other pages illustrate the manner in which 
the various levels were laid off. Tunnels were driven across the 
crater at De Beers mine from west to east, about 120 feet apart, 
and galleries 18 feet wide and 18 feet high were opened every 36 
feet along the main tunnels, and were worked up to within 12 
feet of the loose ground in the top levels. Pillars of solid blue 
ground 18 feet thick were left between the galleries, but later 
on first the roof and then the pillars were taken out. 

This method of mining was fairly successful for a time ; 
but, as already stated, as depth was attained, the roofs of the 
galleries or rooms became unsafe before the galleries were 
opened through to those on the next level above, and they fre- 
quently gave way, thus making the extraction of the blue ground 
exceedingly difficult. This system was both expensive and 
dangerous. No timber was used except in the main tunnels or 
drifts, the nature of the blue ground being such that the roofs 
and sides of the excavations stood fairly well for a short time, 
provided they were well ventilated. 

In other parts of De Beers mine various companies were 
working or trying to work underground ; but as no regular sys- 
tem of mining could be carried on owing to the irregular shape 
of their holdings, and the more or less temporary methods 
adopted, it was clearly impracticable to devise and carry into 
effect any comprehensive system of operation for the rapid and 
economical handling of the diamond-bearing breccia in the 
craters, until the union of all the claims through the formation 
of one controlling company permitted the installation of a single 
uniform system of mining. 

It has already been narrated how this was effected for 
De Beers mine during the year 1887, by the combination of all 
the holdings in the mine into one company, and the organiza- 
tion of De Beers Consolidated Mines Limited, in March, 1888. 
Kimberley mine came formally into the possession of this great 
corporation on the ist of June, 1889, and controlling interests 
in the other two mines, Dutoitspan and Bultfontein, were also 
secured. The assured control of all the mines and their opera- 



SYSTEMATIC MINING 315 

don by De Beers Consolidated Mines Limited enabled its 
directors to institute and conduct successfully a single broadly 
comprehensive plan for extracting the diamond-bearing rock 
and for disposing to the best advantage the total product of 
their mines. 

This system of mining was devised and applied by me 
shortly after my appointment as general manager of the 
De Beers Consolidated Mines Limited, and was based essen- 
tially on a method suggested by the miners themselves and 
without reference to any other system. Instead of attempting 
to withstand, even for a time, the pressure of the superincum- 
bent mass of broken reef, the new system contemplated was a 
caving in and a filling of the excavations, after the precious blue 
ground had been extracted. 

In order to make the output of diamond-bearing ground as 
great as possible, the levels in De Beers mine were at first 
opened up in the new system according to the following plan : 

When the numerous small tunnels had been driven to the 
margin of the mine, i.e. to the point where they reached the sides 
of the crater, the blue ground was stoped on both sides of and 
above each tunnel until a chamber was formed extending along 
the face of the rock for 100 or more feet, with an average width 
of about 20 feet, and about 20 feet high. The roof of the cham- 
ber or gallery was then blasted down or allowed to break down 
by the pressure of the overlying mass of broken diamond-bear- 
ing ground or debris. I mention diamond-bearing ground here, 
for in the early stages of underground mining there was an 
enormous amount of this ground which had been left behind 
when open mining was discontinued, and had been crushed 
either by the moving sides of the immense opening or by the 
collapse of the underground pillars when mined by the old sys- 
tem. It happened frequently, after breaking through to the 
loose ground above, that clean diamond-bearing ground would 
run down as fast as it was removed for weeks or months at a 
time. The galleries would at times become blocked with large 
pieces of blue ground, which had to be blasted, and then a 



316 THE DIAMOND MINES OF SOUTH AFRICA 



further run of blue ground would follow. When the blue 
ground was worked back toward the centre of the crater, larger 
boulders or fragments of basalt, which had come down through 
the loose reef from the surface, would be met with. This sys- 
tem of working would be continued until reef alone came 
down, the waste or reef removed being sent to the surface by 
itself and dumped on the reef tips ; it formed, however, only an 
inconsiderable proportion (one to four per cent) of the total 
output. It will be remembered that, when the roof caved in, 
the gallery was nearly full of blue ground. By the work which 
followed, only a part of this ground was removed by the men 
working on that level, the miners preferring to take it out on the 
next level below. This process of mining was repeated from 
level to level until finally there was no more loose ground to be 
recovered. The cost of extracting blue ground while loose 
ground existed was very low. 



PLAN OF DE BEERS MINE 

8OO FT, LEVEL 




VICTORIA SHAFT 



Now all this has changed, and the plan of opening up new 
levels has altered somewhat, but the system remains the same. 
By referring to the plan, given above, it will be seen that the 



> d *--- "* > TK'"\ " 

* *S*~ : :- \ kF*t \. 

XJ ^^^^ 




SYSTEMATIC MINING 317 

levels were opened around the east end of the mine. When the 
underground works had reached the depth of 800 feet or more, 
a new danger appeared. It will be borne in mind that the huge 
open mines are filled with debris from the sides, caused by the 
removal of the diamond-bearing ground by open quarrying to 
depths varying from 200 to 500 feet. As the supports were 
removed, the sides caved and filled the open mine. This debris 
was composed of the surface red soil, decomposed basalt, and 
friable shale, which extended from the surface down to a depth 
of about 300 feet. In addition to the debris from the surround- 
ing rocks there were huge masses of "floating shale," resembling 
indurated blue clay more than shale. Large heaps of yellow 
ground and tailings, which the early diggers deposited near the 
margin of the mines, and west-end yellow ground contributed 
to the mud-making material. The black shale which surrounds 
the mines disintegrates rapidly when it falls into them. It con- 
tains a small percentage of carbonaceous matter, and a large 
amount of iron pyrites. When the huge masses of shale fell 
into the open mine, they frequently ignited, either by friction or, 
more probably, by spontaneous combustion, as they have been 
known to do on the reef tips, and burned for months and 
even for years at a time. These masses of burned shale become 
soft clay and form a part of the mixture which fills the open 
crater. This debris moves down as the blue ground is mined 
from underneath it, and becomes mixed with the water which 
flows into the open mine from the surrounding rock and with 
storm water, and forms mud. This overlying mud became a 
menace and danger to the men working in the levels below. 
Frequent mud rushes occurred suddenly, without the least warn- 
ing, and filled up hundreds of feet of tunnels in a few minutes, 
the workmen being sometimes caught in the moving mass. It 
became evident that the method of working shown on the plan 
was dangerous in case a mud rush took place, the men being 
sometimes either shut in or buried in the mud coming from 
the opposite side of the mine. It was decided, therefore, to work 
the mines from one side only, and to have the offsets to the 



3i8 THE DIAMOND MINES OF SOUTH AFRICA 

rock connected one with the other at as few points as would still 
allow the ventilation of the working faces. The plan illustrated 



PLAN OF DE BEERS MINE 
1OOOFEET LEVEL 




in the above figure shows the method which was then adopted 
and is still in use. Kimberley mine is worked on about the 
same general system. 




SYSTEMATIC MINING 



3'9 



The method of laying out the workings is also here shown. 
Main tunnels are driven across the crater upon its longer axis, 




Sloping. 



and, at right angles from these, small tunnels are driven out 
every 30 feet until they reach the hard rock on the south side of 
the mine. These tunnels are widened, first along the rock until 
they connect one with another, and, at the same time, the roofs, 
or " backs," are stoped up until they are within a few feet of the 



SKETCH SHOWING METHOD OF STORING 




PLAN I 




SECTION I 




SECTION 




STORES CONNECTED 



320 THE DIAMOND MINES OF SOUTH AFRICA 

loose ground 
above, thus 
forming long 
^ galleries, filled 
more or less with 
blue ground, 
upon which the 

men stand when drilling holes in the backs. The working levels 
were at first 30 feet apart vertically, but, for greater economy, 





Method of Sloping, Vertical Section. 



SYSTEMATIC MINING 



321 



the distance was soon changed to 40 feet. The broken blue 
ground lying in the galleries is taken out, as a rule, before there 
are any signs of the roof giving way. At times this is impos- 
sible, and the roofs cave upon the broken ground, and the blue 
ground is covered with reef. As the roofs cave or are blasted 
down, the blue ground is removed, and the loose reef lying above 
it comes down and fills the gallery. Tunnels are often driven 




Timbering Tunnels. 

through this loose reef, and the blue ground, which has been cut 
off and buried by debris, is taken out; but it is sometimes left 
for those working the next level below to extract. 

After the first " cut " near the rock is worked out, another 
cut is made, and in this manner the various levels are worked 
back, the upper level in advance of the one below, forming ter- 
races as shown in section on page 320. In De Beers mine there 
are now eleven levels on which work is progressing, commencing 
at the depth of 800 feet and extending down to the 1 2oo-foot 



322 THE DIAMOND MINES OF SOUTH AFRICA 




Timbered Tunnels at the zooo-foot Level, Kimberley Mine. 

level. At Kimberley mine nine levels are being worked, from 
the I2oo-foot to the 1520-1001 level inclusive. The galleries are 
not supported in any way by timbers, but all tunnels in soft blue 

ground are timbered 
with sets of two props 
and a cap of round 
timber, and are cov- 
ered with inch and a 
half lagging. 

Soft blue ground 
is drilled with long 
jumper drills sharp- 
ened at both ends. 
In hard blue ground, 
drills and single-hand 

Natives drilling, De Beers Mine. hammers are USed. 

The native workers become very skilful in both methods of 
drilling, and do quite as much work as white men would do 
under similar conditions. 




SYSTEMATIC MINING 



323 



Winding Shafts 

The grand winding shafts and plant by which the enormous 
output of diamond-bearing ground is brought to the surface are 
illustrated in accompanying figures. The present working shafts 
are all vertical. De Beers rock shaft was the first large vertical 
shaft of any importance, from the present mining point of view, 
which was sunk in any of the mines. It is 20 feet by 6 feet in 
size inside timbers, and contains four compartments, two for 
skips lifting blue ground, one for a cage for taking men and 
material up and down, and one for pumps and ladderway. A 




Detail of Sets for Rock Shaft. 

balance weight for the cage runs in the pump compartment, 
which is also the downcast shaft through which the whole mine 
is ventilated. 

No. i is the upcast shaft. It has two compartments for 
skips, two for cages, one for pipes, etc., and a double ladderway. 

At Kimberley mine the rock shaft is a duplicate of De Beers 
rock shaft, except that the pump compartment is larger. 

At De Beers, tunnels 1 1 feet wide by 8 feet high have been 
driven from the rock shaft at the 800, 1000, 1200, 1400, and 
1720-foot levels, and from No. i shaft at the 380, 800, and 
i4OO-foot levels. 



324 THE DIAMOND MINES OF SOUTH AFRICA 

At Kimberley mine the rock shaft is connected by similar 
tunnels with the mine on the 1000, 1200, 1520, 1840, and 2160- 
foot levels. The present depths (1902) of De Beers shafts are 
1720 feet and 1400 feet respectively, and of Kimberley rock 
shaft 2160 feet. 

Trucks holding 16 cubic feet transport the blue ground in 
the mines from the loading places to the main chutes or passes, 
and from these to the shaft. The trucks are hauled by an 




A Shaft Station. 



endless chain which rests upon V-shaped clips fastened to the 
trucks, the motive power being supplied by engines driven by 
compressed air, carried through pipes from the surface. At the 
shaft there is a large station cut out of the solid rock, some 30 
feet wide, and extending back toward the mine for a distance of 
70 feet to the point where the tunnel (8 by 1 1 feet) commences. 
There is an extension on one end of the shaft for a small cage- 
way to bring up any ground that may spill over the skips while 
being loaded. This prevents delays in the skip hoisting. The 
shaft is also lengthened for a few feet at the pump end, where a 
set of pumps is put in. 



SYSTEMATIC MINING 



3 2 5 




As one descends the shaft in the cage in pitch darkness and 
suddenly comes to a large opening brightly lighted with numer- 
ous electric lamps, the scene is weird and confusing. A score of 
natives, half dressed, each vying with the other in shouting his 
own comments upon the visitors as they come forth from the 
cage ; the whirl of heavy iron trucks as they go to and fro ; the 
banging of the tippers as they turn over and deposit the contents 
of the enclosed truck into a chute below, all present a picture 
unique in itself and only to be seen in passing through the shafts 
at De Beers and Kimberley mines. Those who have travelled 
through the native centres, or have seen the negroes loitering 
about the towns, and have thought them lazy, indolent, beer- 
drinking beings, should visit the diamond mines, and especially 
the scene upon the "flat sheet" as described above, and they 
will get a new impression of the working capacity of these 
despised black men. The natives working in the diamond 



326 THE DIAMOND MINES OF SOUTH AFRICA 

mines, if they are old hands in the service, are uniformly active 
and industrious men, while natives fresh from the kraals are soon 
taught their duties, which they learn to perform with nearly as 
much skill as most European miners. 







INVERTED PLAN 

PLAN OF SKIP FOR SIX LOADS. 



328 THE DIAMOND MINES OF SOUTH AFRICA 




No more rapid handling and extraction of the blue ground 
seems possible than is effected by the aid of these alert workers 
and the perfected mechanical devices. As soon as the loaded 
trucks reach the shaft, they are tipped into loading chutes hold- 
ing six truck-loads (96 cubic feet). As the skip reaches the 
bottom a door is opened, and the contents of the chute run into 
the skip and are hoisted to the surface. Experience has shown 
that the best results are obtained by sending up loaded skips 
from one level at a time. The simple and efficient device early 
adopted in the mines for tipping the loads from the trucks into 
the skip at No. i incline of De Beers consisted of an iron 
chute. Four end-tipping trucks were placed close against the 
edge of the chute and the catches loosened. As soon as an 
empty skip was lowered past the chute the trucks were tipped 
and the loads ran into the chutes so rapidly that the engine- 
driver frequently received a signal to hoist before his engine had 
been stopped. The skip in this incline held 64 cubic feet, or 
four truck loads weighing 1600 Ibs. each. 



SYSTEMATIC MINING 



The time of the journey through the shaft now varies only 
a little with depth, being from thirty-five to forty seconds from 
the i2OOor 1520-foot levels. On reaching the surface, the blue 
ground is tipped automatically from the skips into loading boxes. 
The "self-dumping" skips in present use were introduced by 
me in 1888, and were made from drawings supplied by the Union 
Iron Works of San Francisco, and are similar to the skips used 
in the mining districts of the Pacific Coast. (On page 327 
are shown the plans for 
the skip and the manner 
of tipping into the surface 
chutes.) From these chutes 
the blue ground is loaded 
into side -tipping trucks 
holding 20 cubic feet each. 
The average weight of the 
blue ground in a surface 
truck is 2000 pounds. The 
trucks used underground 
hold 1 6 cubic feet, and are 
end-tipping in the inter- 
mediate levels where the 
ground is dumped into 
passes, but have solid ends 
on the main levels where 

revolving tippers are Used. The Rock Shaft, De Beers Mine. 

From the depositing surface boxes at the winding shafts, the 
ground is taken by means of an endless wire rope haulage to the 
" floors," where it is treated as described in another chapter. 

Record Hoisting 

With alert and orderly handling of the blue ground in the 
mines, the rapidity of extraction has advanced to extraordinary 
record points. During the month of July, 1889, 142,567 
loads were hoisted through a single shaft in No. 2 incline, 
De Beers mine. The best day's work of 24 hours was 6222 




330 THE DIAMOND MINES OF SOUTH AFRICA 




VERTICAL TANDEM COMPOUND CONDENSING WINDING ENGINES 



loads of 1 6 cubic feet, or 4977 short tons. For an hour at 

a time hoisting was carried on at this shaft at the rate of five 

skip-loads every three minutes, or 400 truck-loads an hour, 

lifted from the 7OO-foot level, 
a distance of 840 feet through 
the inclined shaft. The total 
amount of blue ground 
hoisted during the fiscal year 
from April i, 1889, to March 
31, 1890, was 1,355,089 
loads, aggregating 1,084,071 
tons of 2000 pounds. This 
remarkable record was made 
under unfavorable condi- 
tions, because the hoisting en- 
gine was small, nominally of 
70 horse-power, and not de- 
signed for such rapid service. 
With the construction of new shafts and the setting up of 

engines and fittings of the 

best and latest designs, the 

efficiency of operation was 

greatly increased. Two types 

of winding engines have been 

erected, and it is interesting 

to follow the changes which 

have been made in this por- 
tion of the plant. The first 

large engine erected by the 

De Beers Company was the 

one at De Beers rock shaft. 

Its cylinders were 24 inches 

in diameter, with a stroke of 

5 feet. It had two drums, 

each 4 feet 4^ inches in width 

and 10 feet 6 inches in diam- 




VERTICAL TANDEM COMPOUND CONDENSING WINDING ENGINES 



SYSTEMATIC MINING 




VERTICAL TANOEM COMPOUND CONDENSING WINDING ENGINES 



eter, with a grooved tread to prevent friction 
on the rope. This engine was built by well- 
known makers of winding engines, whose 
works are too near the cheap coal centres 
of England. The engine was what is 
called in America a "sawmill engine." 
In the timber districts of America, the 
boilers are fired with slabs cut from the 
round logs in squaring them. Enor- 
mous quantities of these slabs accu- 
mulate about the mills, where they 
must be consumed in some way, or 
carted away at a considerable ex- 
pense. To get rid of the slabs 
engines that consume the 
greatest amount of steam 
are those most sought after. 
In South Africa, on the contrary, the extraordinary consumption 
of steam was a heavy drawback. Welsh steam coal then cost 
_8 ioj. ($41.25) per ton of 2000 pounds, delivered at Kim- 
berley, so this "sawmill" engine was converted from two high- 
pressure cylinders to a cross compound, with cylinders of 26 
inches and 40 inches diameter, and the consumption of fuel 
dropped more than 30 per cent. After several years of con- 
stant service, the engine was stopped June n, 1896, the old 
drum and crank shaft, weighing 32 tons, were taken out bodily, 
and a new set, weighing 40 tons, substituted and made ready 
for service in less than 48 hours. (See illustrations.) With 
this new outfit there was soon a series of record-breaking per- 
formances, which are given below. 

At the Kimberley mine, the main or rock shaft was started 
on the north side of the mine in March, 1889. I" tne & rst Y ear 
this shaft was sunk to the depth of 699 feet, and, in the following 
year, it was pushed to the depth of nearly 1300 feet. The driv- 
ing of the tunnel to the mine from this shaft on the looo-foot 
level showed how exactly vertical was the wall of the crater, 



332 THE DIAMOND MINES OF SOUTH AFRICA 




\Vinding Engine, Kimberley Mine. 

for the tunnel, at this depth, entered the blue ground 1134 feet 
from the shaft, corresponding almost precisely to the distance 
from the mouth of the shaft to the edge of the melaphyre at the 

depth of 300 feet. For hoisting 
service at this shaft, a winding-en- 
gine plant was especially designed 
by the late Mr. Louis I. Seymour, 
mechanical engineer for De Beers 
Consolidated Mines Limited, and 
constructed by James Simpson & 
Co., of London, England. This 
plant consisted of a pair of vertical 
tandem compound engines driving 
two reels. These engines were de- 
signed to hoist six truck-loads in 
one skip, from the looo-foot level, 
in 45 seconds, including filling, 
Mr. Louis i. Seymour. starting, stopping, and discharging; 




SYSTEMATIC MINING 



333 



but in practice they pulled up the skip carrying this load 
from the icoo-foot level in from 30 to 35 seconds. Flat ropes 
were used, at first, on the reels, but when the shaft was sunk 
some hundreds of feet deeper, round ropes were substituted by 
the adoption of the " Whiting system," first used by Mr. S. B. 
Whiting, general manager of the Calumet and Hecla Copper 
Company of Michigan. The dimensions and description of the 
engines are given in Appendix II. 

tE-BEERS CONSOLIDATED MINES LTD. 



PLAN OF 600 FT. LEVEL 

BULTFONTEIN MINE 




The only excuse I can offer for having adopted flat ropes 
for winding is that I was persuaded to do so against my own 
judgment by a number of American engineers, and experience 
proved that I erred in so doing. Leaving all other disadvan- 
tages aside, and they are many, the extra cost of ropes per load 
is sufficient to condemn the flat rope. The average cost per load 
for flat ropes was .6 of a penny against .076 of a penny with the 



334 THE DIAMOND MINES OF SOUTH AFRICA 

present Whiting system, the saving amounting to more than 
j2ooo per annum. This system as modified in the diamond 
mines is as follows : The round winding rope, made of the best 
plough steel, extends from the skip over the sheave on the pit- 
head frame down to the reel on the crank shaft of the engine, 
thence four times around this reel and a corresponding reel on a 
lay shaft (centres of shafts being 12 feet apart) ; thence the rope 
passes around an idler sheave, the shaft of which runs on bear- 
ings set upon a movable frame, which is attached at each end to 
a carriage by means of trunnions. The carriage in this case 
runs upon a track 50 feet long. From the idler or tension 
sheave the rope passes around a second reel which is loose 
upon the crank shaft, the centre of which is in line with the 
second sheave upon the pit-head frame. 

By the completion of the new plant the output of blue 
ground from the Kimberley mine was greatly increased. Dur- 
ing the fiscal year ending June 30, 1893, I >453> 1 5 2 loads were 
taken from the mine as against 1,310,994 loads, the output 
for fifteen months previous, an increase almost wholly due to 
the new hoisting facilities, for fully three-fourths of the yield 
was drawn through the main shaft. The product of De Beers 
mine for the same year was still greater. The total quantity 
hoisted was 1,637,031 loads, of which 1,403,060 loads were 
drawn through the main or rock shaft, and only 233,971 loads 
through the No. 2 or west end incline shaft. 

Drainage 

Thorough drainage is of manifest importance in the opera- 
tion of any mine, but it is peculiarly essential in these diamond 
mines. At the commencement of underground mining the 
inflowing water was removed by steam pumps. The use of 
such pumps was an error, for the resultant heat and moisture 
caused the blue ground to crumble, and made the ladderways 
so hot that they were at times impassable. 

As soon as the vertical shafts were completed at De Beers 



SYSTEMATIC MINING 335 

and Kimberley mines, Cornish pumping plants were put in, by 
means of which all the water is now pumped from the mines. 
The average quantity of water taken from De Beers mine is 
435 gallons per diem, and from Kimberley, 8385 gallons. 
Nearly half of the latter influx comes from a crevice at the junc- 
tion of the quartzite with an intrusive dike of igneous rock 
which was struck while driving the iioo-foot tunnel at a dis- 
tance of 600 feet from the mine. While no water is found in 
the blue ground or mine itself, that which flows into the mine 
from the surrounding rock mixes, as before described, with the 
debris which has fallen into the worked-out portion of the 
De Beers and Kimberley mines, and makes mud. Enormous 
quantities of this mixture are from time to time forced suddenly 
into the working parts of the mine, which are connected by tun- 
nels with the loose debris. At times hundreds of feet of tunnels 
were filled in a few minutes. Mud rushes became so frequent 
that the working of the mines was seriously interfered with, and 
the loss of life was very great. 

At Kimberley mine, large springs of water flowed into 
the open works at the junction of the melaphyre with the 
shale. Only a small part of the melaphyre was then exposed 
to view, and the position of the other part was unknown. 
A tunnel was started from the Standard shaft, and driven to 
the south around the mine. Another tunnel was started from 
the Harvey shaft and driven to the west end around the mine 
in the opposite direction until the two tunnels met. The 
total length was 2097 feet. Through these tunnels all the 
surface water and all water coming into the mine above the 
melaphyre was taken up and led to the pumps by means of 
pipes. All water which enters the mine in the deeper work- 
ings is taken down in passes, sunk in the rock outside of the 
mine. By these precautions mud rushes have been completely 
stopped in Kimberley mine, and none have occurred for many 
years past. 

De Beers mine has not been so fortunate, and mud rushes 
are of frequent occurrence, although the quantity of water in 



336 THE DIAMOND MINES OF SOUTH AFRICA 

this mine is only about one-half that of the Kimberley mine. 
The following work is being done with the view of preventing 
them. A tunnel is being driven around the mine at the hard 
rock (melaphyre) level, about 380 feet from the surface, in 
order to take up the water which flows into the open mine 
below the shale. Tunnels are also being driven on the 1000- 
foot level on the south and east sides of the mine, which will be 
continued until they meet. Diamond drills are at work making 
holes between levels, with the view of tapping the water. 
Everything feasible will be done to free De Beers mine from the 
plague of water as perfectly as it has been done at Kimberley 
mine. The problems are not the same, however, for in Kim- 
berley mine the debris had followed down as the blue ground 
was extracted, and had left the hard rock more or less exposed 
to view, and one could see in places where the streams of water 
flowed into the open mine ; but in De Beers no hard rock has 
been exposed until lately, and one must grope in the dark, as it 
were, to find out where the water enters the open or worked-out 
portions of the mine. 

The pumping plants for freeing the mines from water have 
kept pace fully with the advance in the hoisting plants. For 
the service of De Beers mine, a new pumping engine was 
erected at the rock shaft in 1889. This is a compound sur- 
face-condensing engine made by James Simpson & Company, 
of London. Its high-pressure cylinder is ''4^4 inches diam- 
eter, and its low pressure, 21 inches, with a stroke of 30 inches, 
It is capable of developing 120 horse-power. With this engine an 
average of nearly 6000 gallons an hour was readily drained from 
the mine from the start, and no difficulty was experienced in lift- 
ing over 8000 gallons an hour at times. The cost of pumping 
is largely offset by using the water drained from the mine for 
washing the pulverized blue ground. By combining this sup- 
ply with that obtained from surface reservoirs, enough water 
was obtained for the use of the concentrating plants, except in 
very dry seasons. For the Kimberley mine a Cornish pumping 
plant of 400 horse-power, from designs by the late Mr. L.I. 



SYSTEMATIC MINING 337 

Seymour, was erected in 1891. This is a vertical triple-expan- 
sion condensing engine, with cylinders 15^ inches, 23^ inches, 
and 37 inches in diameter, and a stroke of 36 inches. The gears 
for this engine were made by Fraser & Chalmers, of Chicago, 
Illinois, and the crank shafts by Sir J. Whitworth, of Manches- 
ter, England, but the main constructors were James Simpson 
& Co. Ltd., of London. (See Appendix III.) With this 
plant an average of over 12,000 gallons a day was readily 
pumped from the mine in the first year after its erection, 
and since then there has been no further difficulty in handling 
the influx of water into the workings. 



Compressed Air 

For all underground service in the mines, in driving sinking 
engines, mechanical haulages, rock drills, and any other machin- 
ery where power is necessary, steam has been supplanted by 
compressed air. Electricity has also been used for some of these 
purposes, and is the cheaper and better power for many of the 
uses for which steam and compressed air have been used. 

Lighting 

For lighting, the application of electricity has already proved 
to be almost indispensable. All tunnels and ladderways through- 
out the mines are lighted by electricity. In the stopes and other 
working faces candles are used. Electric lights have been found 
to be of the greatest assistance in enabling the men to get away 
from rushes of mud. These occur at times when some of the 
galleries are " hung up " (to use a miner's expression), which 
means when the tops of some of the galleries are choked with 
huge pieces of blue ground. The roof suddenly gives way from 
the pressure of mud above, and all open lights, such as candles, 
are put out by the force of the concussion of the air, and, were 
it not for the electric lights, the tunnels in the vicinity of the 
mud rush would be in total darkness. 



338 THE DIAMOND MINES OF SOUTH AFRICA 

Other Electric Service 

Electric bells are in use throughout the mines, and have 
very greatly promoted the rapidity of hoisting through the 
shafts. Owing to instantaneous communication between the 
man in charge of loading the skips and the engine-driver, hun- 
dreds of loads more are sent to the surface daily than could 
be forwarded under the old " pull bell " system. For instant 
additional communication between the surface and the under- 
ground work telephones have been installed, and the same rapid 
communication extends to the depositing floors, concentration 
works, and offices of the company. 

Natural Ventilation 

The Kimberley mine is ventilated in a somewhat peculiar 
manner. The rock shafts at both De Beers and Kimberley 
mines are downcast, i.e., the air for ventilation goes down these 
shafts, along the bottom tunnels and thence up through the 
various levels, and it is fortunate for the men working in the 
mine that it is so, for the cool air comes in at the bottom and 
ventilates the mine much better than if the rock shaft drew the 
heated air down through all the lower workings. The upcast 
in Kimberley mine is through the Harvey shaft, the top of 
which is 300 feet below the top of the rock shaft. This shaft, 
with which the various levels of the mine are connected, extends 
down to the I2oo-foot level, and a similar shaft or winze situated 
near it extends from the I2oo-foot to the levels below. As the 
top levels in the mine are the hottest, the current of air ascends 
through the Harvey shaft. The usual direction of air currents 
in mines with two shafts and natural ventilation is down the 
shorter shaft and up the shaft the mouth of which is situated at 
the greater height on the surface. The reverse is the case at 
Kimberley mine. The quantity of air which passes down 
De Beers rock shaft was 33,300 cubic feet per minute until 
1898, when the enlargement of the upcast shaft was completed, 



SYSTEMATIC MINING 339 

and the air current was increased to 45,000 cubic feet per 
minute. In the Kimberley mine the influx of air per minute 
is 25,500 cubic feet. 

Temperatures 

At De Beers, with temperature of the air on the surface 
79 F., the temperature ranges from 75 to 77 in the tunnels 
leading to the mine on the 1000, 1200, and i4OO-foot levels. 
The temperature of the air as it leaves the mine on the 8oo-foot 
level is 84. The temperature of the mud after a mud rush 
was on one occasion 85 F. Temperatures at Kimberley mine 
in the i2oo-foot tunnel were, for the air, 71. 5; for the rock, 
72. i ; for the large spring of water 78. 9 F. The quantity of 
water flowing from this spring, which is about 600 feet from the 
crater, is 3500 gallons an hour. The temperature in the work- 
ing galleries on this level is 87. Springs of water on the 1520 
and i84O-foot levels gave 83. 8 and 8i.9 respectively, the 
water in the lower level being the cooler. 

The Output of Slue Ground 

The table of statistics (Appendix V) gives the amount of 
blue ground produced from De Beers and Kimberley mines since 
the formation of the De Beers Consolidated Mines Limited. 
In the same table of statistics will be found the average cost 
of production per load. It will be seen that the lowest cost was 
for the year ending June 30, 1894, 6s. 6.8d. per load. This 
includes all charges from the mining of the ground to the 
delivery of the diamonds to the valuators. All mine charges, 
including shaft-sinking, tunnelling, etc., are charged to current 
expenses. It is interesting to note in the same table the great- 
est output of each mine through a single shaft for various 
periods of time. The maximum amount of blue ground pro- 
duced in one year was 1,746,240 loads from De Beers mine for 
the year ending December 31, 1897. This ground was raised 
from a depth of 1000 feet from the beginning of the year until 
June 14, when winding commenced from the isoo-foot level. 



340 THE DIAMOND MINES OF SOUTH AFRICA 

At Kimberley mine, the maximum output from one shaft was 
1,600,422 loads for the year ending December 31, 1893, hoisted 
from a depth of 1000 feet. These figures do not include the 
waste or "reef" which is taken out, amounting to 100,651 loads 
from De Beers mine and 64,799 loads from Kimberley mine 
during the year. 

During the month of November, 1898, 208,013 loads of 
blue ground equal to 166,410 tons of 2000 pounds, were hoisted 
through the two skip compartments of De Beers rock shaft and 
from a depth of 1200 feet. The winding stops from Saturday 
night at eleven o'clock until Monday morning at six. The 
average number of loads of blue ground hoisted per hour was 349. 
The average daily output for a full day's work was 8376 loads, 
and for Saturdays 5933. The best day's record was 9790 loads, 
the best week's record was 50,450. In the above records no ac- 
count has been taken of stoppages during working hours nor is 
the quantity of waste, which was 11,992 loads during the month, 
taken into account. Previous to this the best month's produc- 
tion was from De Beers mine, in November, 1897, a total of 
197,173 loads from the I2oo-foot level. In Kimberley mine, 
the best records for a month were in November, 1893, when 
157,847 loads were taken from the looo-foot level, working 
three shifts of eight hours each per day, and 108,627 loads from 
the i2OO-foot level in May, 1895, working twelve hours per day. 

The best week's record from Kimberley mine, winding by 
day only, was 27,418 loads in sixty-nine hours from the 1520- 
foot level for the week ending September 22, 1897. No account 
has been taken of any lost time. 

From the above figures it will be observed that all records 
have been broken for winding ground through a single shaft with 
two skip compartments. 

Labor and Wages 

The following table shows the average number of men 
employed in and about the mines worked by De Beers Consoli- 
dated Mines Limited, during the year ending June 30, 1897 : 



SYSTEMATIC MINING 



EMPLOYES 
The average number of persons daily employed is as follows : 





DE BERKS. 


KlMHfcRLEY. 


PREMIER. 


WORKSHOPS. 


ON THE ESTATES 
AND ELSEWHERE. 


Whites. 


Blacks. 


Whites. 


Blacks. 


Whites 


Blacks. 


Whites. 


Blacks. 


Whites. 


Blacki. 


Above ground 
Underground 


477 
212 


1851 
2OOI 


187 
83 


925 
1322 


46 

>5 


423 

489 


388 


211 


28 


118 



TOTAL. 



Above ground 
Underground 



Grand total . 



Whites. 


Blacks. 


1126 
500 


3528 
3812 


1626 


7340 



The average number of white men employed has increased to over 2000 and the num- 
ber of natives to over 11,000. 

NATIONALITIES OF WHITE EMPLOYES 
PERCENTAGES 





IN AND ABOUT THE 
MINES AND FLOORS. 


AT THE WORKSHOPS. 










1894. 


i8 97 . 


,894. 


1897- 


English ........ 


52.2 


46.5 


41.5 


37-' 


Scotch ........ 


6.2 


7-2 


23-3 


2O.6 


Irish 


4.8 


5.6 


2.4 


2.8 


South Africans ...... 


33-i 


36.8 


27.1 


33-3 


European ....... 


1.8 


'5 


4.2 


4-7 


Other Nations ...... 


1.9 


24 


'5 


'5 




100 


100 


IOO 


IOO 



WAGES 

The following figures give about the average wages paid for various kinds of 
labor at the mines : mechanics, ^5 to j per week ; miners, from ^5 to 6 
per week ; guards and tallymen, from $. to $ per week ; engine-drivers, ^6 
to -j per week ; natives in the underground works, from y. to $s. a day. 

Overseers, from ^3 1 2s. to ^4 zs. ; machine men and assorters, from _-$ 
to 6 ; natives (ordinary laborers), ijs. 6d. to 2 is. per week; drivers, from 
2$s. to zjs. 6d. per week. Every employe has a percentage on the value of 
diamonds found by himself. On the floors the white employes receive is. 6d. and 
the natives ^d. per carat. Nearly double these amounts are paid for stones found in 
the mines. 



342 THE DIAiMOND MINES OF SOUTH AFRICA 

Dutoitspan and Bultfontein Mines 

If operations were not pushed with the like energy and lib- 
erality of outlay in Dutoitspan and Bultfontein mines, it was 
simply because of sound economic considerations, and impedi- 
ments unreasonably placed in the way of projected developments. 
Heavy falls of reef had very greatly damaged the open work- 
ings in Dutoitspan mine before it came into possession of the 




Mount Ararat before Blasting. (Removal of a piece of " Floating Reef," Bultfontein Mine, 1901. 
It was 150 feet long, 50 feet wide, and 120 feet high. 180 holes were drilled in it and charged 
with 1050 pounds of No. i dynamite.) 

De Beers Consolidated Mines. In spite of this obstacle, work 
was carried on actively for a time, until it became certain that no 
profit could be made by working this mine and the continuance 
of operations would have caused great loss directly to the con- 
trolling corporation. If diamonds were like gold and there was 
an unlimited demand for the product, Dutoitspan mine would 
assuredly have been worked as long as it paid expenses and the 
barest margin of profit. But, seeing that the demand for dia- 



SYSTEMATIC MINING 



343 



monds, or any other precious stones, is practically limited to the 
amount marketable without breaking down the prices dis- 
astrously to the producer as well as to every dealer and cutter, 
work in Dutoitspan mine was suspended at the close of 1889. 
The mine is still idle, but a large shaft was started in 1901 for 
the purpose of working it at some future date. 




Shots Fired. 

Bultfontein mine might have proved more profitable, but in 
1889 an immense fall of reef, covering nearly the whole bottom 
of the mine, made open work impossible, except over a very small 
area. In face of this situation shafts would have been started 
and underground work on a systematic plan prosecuted, had it 
not been for obstacles set in the way by the lessors, the London 
and South African Exploration Company. It was not antici- 
pated that there would be any profit in instituting these costly 



344 



THE DIAMOND MINES OF SOUTH AFRICA 



underground works at that time, but the directors of De Beers 
mines desired to furnish employment to miners out of work, 
and the mine would have been opened and explored on a return 
of bare expenses, if the lessors had seen fit to make reasonable 
terms. As their demands were considered exorbitant, work in 
this mine was also stopped in 1889, and was only commenced 




A Second after Firing. 

again in 1900. Plan on p. 333 shows how the mine is being 
opened. There are nearly 13,000,000 loads of blue ground 
in sight above the 6oo-foot level. 

Premier Mine 

In December, 1891, the farm, Benaauwdheidsfontein, adjoin- 
ing Kimberley, and lying on the border line between Griqualand 
West and the Orange Free State, was purchased in full by the De 



SYSTEMATIC MINING 345 

Beers Consolidated Mines. On this property the Wesselton or 
Premier mine, situate about four miles from the town of Kimber- 
ley (plan at pages 316-317 gives its position relative to the other 
mines), had been discovered in September, 1890, by a Dutchman, 
Fabricius, who was prospecting for an old resident of the dia- 
mond fields, Mr. Henry A. Ward, who had a bond on the Wes- 
sels' estate, or an option to purchase the property for ^175,000 
within a stated period of time. When a man has no money, 




The Mine Filled with Smoke. 

and Ward had little or none at that time, it matters very little 
to him what amount he has to pay for such a property, for he 
does not want the farm unless he finds a payable diamond mine, 
and if he does find a mine, some one else supplies the funds. In 
this case the mine was found, but it was one chance in a million. 
Only a small portion of Wessels' farm was in the Cape Colony, 
and it was upon this portion that the mine was discovered. 

Scores of sanitary pits had been sunk within a stone's throw 
of the mine before the prospector Fabricius sunk a hole at ran- 



346 THE DIAMOND MINES OF SOUTH AFRICA 

dom, without any apparent reason, through ten feet of limestone 
and found yellow ground. It was soon noised about, and the 
mine was rushed and jumped by a crowd from Kimberley and 
Beaconsfield, consisting to a great extent of members of the 
Knights of Labor. Hundreds of claims (30 feet by 30 feet) 
were pegged off, and holes averaging 3 feet by 6 feet were sunk 
all over the place, looking far more like open graves than pros- 
pectors' shafts ; in fact, they proved to be the graves of the 




After the Smoke has cleared away. 

hopes of the reckless jumpers of private property. Many of 
the holes were sunk outside of the area of the present mine. 

Ward had the sole right of prospecting for minerals upon 
this farm, which was held under his agreement with Wessels; 
but for some time the jumpers held their ground regardless of 
its legal ownership, and their contest was the more bumptious 
from the fact that the mine was only a few hundred yards from 
the boundary line between the Colony and the Free State. 
Title to Wessels' farm was originally granted by the Free State. 



SYSTEMATIC MINING 347 

By the laws of this State all minerals belong to the owners of 
the farms upon which they are found. In the settlement of the 
boundary line between the Free State and Griqualand West it 
was agreed that the farmers who had held titles to their farms 
under the laws of the Free State should retain the right to any 
minerals that might be found upon them. After months of 
wrangling, Ward's claim was established beyond dispute. Ward 
was without means to continue prospecting, and parted with half 
his rights for ^3,000. When the mine was discovered, De Beers 
Consolidated Mines bought the interest which Ward had sold, 
for which they paid ^120,000. Ward disputed De Beers owner- 
ship to an undivided one-half interest in the property. The 
case came to trial in the Supreme Court of the Cape Colony, 
the mine having been discovered in that part of the farm lying 
within the Colony. Judgment was given in favor of De Beers, 
and that Company became joint owner with Ward in the prop- 
erty, now called the Premier Mine, named by Ward in honor of 
Rhodes, who was at the time Premier of the Colony, and with 
whom he had conducted most of the negotiation in relation to 
the purchase of the mine and the final disposition of his interest. 
In the meantime Ward had obtained an extension of his 
option for an additional sum of ^125,000. The directors of 
De Beers mines were in no way consulted in this matter. The 
time for taking up the option was drawing to a close, and as 
Ward did not have the money to pay for his half, it was mutually 
agreed that De Beers should pay the purchase price of ^300,000, 
Ward becoming responsible for the repayment of his half. 
After considerable negotiation Ward agreed to cede his interest 
in the mine on the following conditions : that he should take 
over the mine for a period of five years, during which time he 
had the right to take out 5,000,000 loads, equal to 4,000,000 
tons of diamond-bearing ground. Diamond-bearing and blue are 
not synonymous terms here, for Ward took out yellow ground 
to the depth of about 60 feet. The mine was surveyed as 
accurately as possible. An allowance of 8 feet in depth was 
made for the surface limestone which covered the mine and 



348 THE DIAMOND MINES OF SOUTH AFRICA 




SYSTEMATIC MINING 349 

which was supposed to be o-diamond-bearing. It was also 
agreed that a load of ground in place should be 9.6 cubic feet, 
but this was afterward increased to 10.6, as it was found that 
9.6 cubic feet of yellow ground would not make a load when 
broken. From the preliminary washing of ground taken from 
various parts of the mine, it was estimated that the mine would 
yield about 16 carats per hundred loads washed. Ward took 
possession of the mine, and through contractors erected a 
large washing plant capable of washing 4,000 to 5,000 loads 
daily. During the five years Ward mined and washed the 
5,000,000 loads to which he was entitled. The yield was about 
20 carats per 100 loads by means of the first sortings, and pos- 
sibly two or three carats more were obtained by subsequent sort- 
ing, so that the total number of carats obtained reached about 
1,100,000. As to the price realized for these diamonds and the 
cost of producing them, I have no knowledge, but one may 
assume that the average value of the diamonds was about i8j. 
per carat, and that the cost of mining and washing did not 
exceed is. 6d. per carat, if it reached that figure. The first 60 
feet were easily mined, as the ground was decomposed and could 
be sent direct to the washing machines from the mine. At the 
present time, under De Beers management, blue ground is mined 
and deposited, harrowed and watered, and then loaded and sent 
to the washing machines for a cost of about is. id. a load. 

From the year 1871, when the four mines at Kimberley 
and the Jagersfontein mine were discovered, a period of twenty- 
one years elapsed during which no paying diamond mine was 
found, although continuous prospecting was carried on. The 
Premier mine was covered for an average depth of eight 
feet with lime, which for the most part was diamond-bearing. 
The formation of the lime seems to have been the result of the 
evaporation of water highly impregnated with lime, or possibly 
springs existed in the localities, whose waters were highly impreg- 
nated with carbonate of lime, which was deposited by the evapo- 
ration of the water. Water, in many of the lime-covered dis- 
tricts, is found very near the surface. On the Wesselton estate 



350 THE DIAMOND MINES OF SOUTH AFRICA 




SYSTEMATIC MINING 




$52 THE DIAMOND MINES OF SOUTH AFRICA 

it shows itself in numerous " fonteins " or springs. Below the 
lime coverlet the diamond-bearing yellow ground extended to a 
depth of sixty feet, where it changed to blue ground. The 
work which had been done proved the area of the mine, and it 
was found to contain about 1162 claims of diamond-bearing 
ground, equal to about 24 acres. 

Under Ward's administration the diamond-bearing ground 
was removed by means of trucks drawn by an endless chain 
haulage, which delivered them at the top of a large washing 
plant, where it was at once treated. 

In January, 1896, Ward's lease expired, and from that time 
work in this mine has been constantly carried on by the 
De Beers Consolidated Mines. An incline with a grade of one 
foot in five was constructed in 1896; the drainage water from 
all parts of the mine was concentrated in a sump, and a pump- 
ing plant erected capable of handling the great influx of water, 
averaging 42,726 gallons an hour in 1896, or about 7,178,000 
gallons a week. At the end of that year the new works were 
so far advanced that 271,777 loads of blue ground had been 
raised. For the economic working of this mine, a complete 
mining and washing plant, with compounds, machine shops, 
stores, and other necessary buildings, was installed soon after 
the mine was turned over to the Company by the lessee. 

The incline mentioned above was made through the marginal 
reef, and down to a depth of 185 feet. At present the diamond- 
bearing blue ground is hauled from the mine by means of an 
endless wire rope haulage (see illustration opposite) driven by an 
engine on the surface. The mine is being worked in sections of 
50 feet in depth. The ground is broken by drilling deep holes 
(12 feet) with jumper drills and blasting with dynamite. The 
average number of loads broken per case of dynamite (50 Ibs. 
net) is 416, equal to 333 tons. The breaking of the ground 
was formerly done by contract, and cost tfad. per load delivered 
upon the " flat-sheets " near the mine end of the wire rope 
haulage. This mining is now done by the Company. Loading 
is done in the mine upon the contract system, by paying the 



SYSTEMATIC MINING 



353 




Premier Mine, looking up the Incline. 

natives i$s. for 100 loads. The cost of hauling and depositing 
is about 6d. per load. In open mining the natives are paid i$s. 
per 100 loads (80 tons) for loading and delivering to a flat- 
sheet from 100 to 150 feet from the place of loading. On 
the floors, after the ground is pulverized, I2s. per 100 loads is 
paid for reloading. The ground is treated in the same manner 
as at De Beers and Kimberley mines, which will be described in 
the following chapter. 

There is a large body of floating reef in the mine, which 
measured about 350 feet by 200 feet on the surface, but, at the 
depth of 500 feet, it has been nearly displaced by diamond-bearing 
ground. As already mentioned, these large blocks of floating 
reef are portions of the country rock which have broken loose 



2 A 



354 THE DIAMOND MINES OF SOUTH AFRICA 

during the time the craters were being filled, and were not 
incorporated in the breccia of which the blue ground is com- 
posed. In some instances the "floating reef," or "islands," is 




One of the Early Washing Machines. 

the same as the amygdaloid rock or melaphyre, which surrounds 
the mines at a depth varying from 300 to 400 feet, but, as a rule, 
somewhat altered. It will be noticed that the aerial gear is not 
used at the Premier mine, and the reason is that for shallow 
depths, or for depths down to 200 feet, inclines, either open cuts 



SYSTEMATIC MINING 



355 



or shafts inclined to as great an angle as it is practicable for a 
wire rope haulage to work, are more economical. 

Up to the present time the difficulties of falling reefs, which 
caused so much trouble in the other mines, have not yet arisen. 
A belt of blue ground some seventy feet in thickness has been 
left standing in places to support the friable decomposed basalt 
and shale with which the mine is surrounded. This is but a 
temporary remedy, and one which does not recommend itself to 




One of the Present Washing Plants. 

the engineer, owing to the value of the ground which is being 
temporarily sacrificed. It is my intention to combine the open 
with the underground system, and to remove the blue ground 
which lies adjacent to the reef in the same manner as it is now 
done in De Beers and Kimberley mines, and at the same time 
to work the remaining portion of the mine in the open, as at 
present, so long as open mining can be safely and economically 
carried on. Owing to the enormous flow of water from the reef 
into the mine (the blue ground itself contains no water), it will 
be necessary to sink a shaft, and to drive tunnels to tap these 
large springs, and lead the water away from the mine. The 



356 THE DIAMOND MINES OF SOUTH AFRICA 




SYSTEMATIC MINING 



357 



average quantity of water pumped from the mine is about 
40,000 gallons per hour, or more than three times the quantity 
which is pumped from De Beers and Kimberley mines com- 
bined. In order to make use of this water, it is pumped to 
De Beers floors for washing the blue ground, and to the village 
of Kenilworth for irrigation purposes. 

The average yield of diamonds for several years past under 
De Beers management has been three-tenths of a carat per load. 




No. 2 Washing Plant, De Beers Floors. 

The value of the Premier mine diamonds as compared with 
those from De Beers and Kimberley mines is about twenty per 
cent less, owing to the greater proportion of boart and small 
diamonds. The diamonds from this mine show distinctive char- 
acteristics, and a parcel of them can be easily distinguished from 
those produced from other mines. It is estimated that the 
production of this mine could be raised to 1,000,000 carats 
per annum. The mine is being developed for the commence- 
ment of underground mining. Plan on page 318 shows the 
shape and size of the mine on the 5oo-foot level. It is 
estimated that there are 13,000,000 loads, equal to 10,400,- 
ooo tons, of blue ground in sight above this level. The 
Premier mine may, therefore, be looked upon as a mine of 
very great value, and one which will play an important part in 
the future history of the diamond-mining industry. 



358 THE DIAMOND MINES OF SOUTH AFRICA 



Jagersfontein 

Mention has been made previously of the Jagersfontein 
mine. It was the first of the so-called " dry mines " discov- 
ered. The mine is very large, containing 1 1 24 claims. The 
average yield of the ground is about twelve carats per one hun- 
dred loads. The quality of the diamonds far surpasses the 
yield of any other crater. The mine is noted for its large blue- 
white diamonds, and, now and again, an exceptionally large stone 
is found. One stone cut as a brilliant weighs 239 carats and is 
without a flaw. 

Two full-size reproductions are here given of the largest dia- 
mond found in the mine, its weight being 969^ carats. For 

many years after their 
discovery, the richer 
mines of Kimberley 
offered greater induce- 
ments to the digger as 
well as to the investor, 
but the fever for con- 
solidation attacked the 
directors of some of the 
principal companies in 
this mine, and the New 
Jagersfontein Mining 
and Exploration Com- 
pany Limited was 
incorporated in 1888, 
about the same time 
as De Beers, and the 
various interests were 
gradually absorbed. 
The mine is still worked in the open, and during the last few 
years has had some difficulty with falls of reef, which is quartzite 
from the surface down. In 1900 the mine had reached a depth 
of 450 feet in the lowest workings. 




Excelsior Diamond, 969} Carats. (Found in Jagersfontein 
Mine. Actual Size.) 



SYSTEMATIC MINING 359 

The output of blue ground for the year ending jist March, 
1899 (the last full year's work before the war) was 2,600,000 
loads, and the diamonds produced amounted to 289,000 carats, 
which realised about ^500,000. The dividends paid during 
the same period amounted to ^ 150,000, equal to fifteen per 
cent on the capital. 




Another View of the Excelsior Diamond. 



CHAPTER XII 



WINNING THE DIAMONDS 




T has been shown how resourceful engineering 
mastered the problem of the extraction of the 
diamond-bearing deposits swiftly and systemati- 
cally, without injury to the mines. It was 
no less essential to advance and perfect the 
process of the winning of the diamond from 
the mass of extracted blue ground with corresponding speed and 
efficiency. For the handling of the mammoth bulk of breccia, 
through which the tiny, precious crystals were sprinkled in a 
proportion so infinitesimal, there was a practical call for every 
feasible stretch of invention in transportation, concentration, 
assorting of sizes, and final separation of the gems. The indis- 
pensable reconciliation of thoroughness in extraction with rapidity 
in working over the ground made the task greatly perplexing. 
It was only through years of experimenting and progressing from 
imperfect to improved designs that the present great diamond- 
winning plant of the mines was evolved. If this is still short 
of ideal suitability to the work, it is simply fair to observe how 
vast is the stride that has actually been made in a few recent 
years in diamond-winning methods, from the primitive Indian 
wooden shovels and drying mats, and the water holes and shak- 
ing plates of the Brazilian. 

As fast as the blue ground is dumped automatically from the 
skips into the ore bins, it is carried away in trucks by an endless 
wire rope haulage, driven by steam, to the depositing floors. 
These floors are made by removing the bush and grass from 
fairly level stretches of ground. After clearing the face of the 
ground, it is hardened and smoothed with heavy rollers until it 

360 



WINNING THE DIAMONDS 



361 



is fit for use. Receiving grounds are laid out separately for each 
of the diamond mines on the four farms, and cover an area of 
several thousand acres. The most extensive of any are the 
De Beers floors, which are laid off in rectangular sections, six 
hundred yards long and two hundred yards wide, on the farm, 
Kenilworth, adjoining the mine. They begin about a mile from 
the mines and extend for three miles in the easterly direction and 
a mile to the west. 

The main tramway line from the mine is three miles in 
length, with two branches, one mile and three quarters of a mile 




De Beers Mine and Floors. (Showing Haulage from Mine to Floors.j 

in length respectively. The speed of the running trucks ranges 
from 2.5 to 4 miles an hour, and they are counted and greased 
automatically as they are sent on to the floors. There is a slight 
down grade from De Beers and Kimberley mines which is of 
material service in lightening the drag of the loaded trucks. 
When the trucks reach the floors, they are drawn by horses or 
mules over auxiliary tram lines at right angles to the main haul- 
age line to any desired point of deposit. A full truck contains 
about 1 6 cubic feet of blue ground, weighing 1600 Ibs. approxi- 
mately; but it was found more convenient to supplant these 
end-tipping trucks by 20 cubic feet side-tipping trucks. The 



362 THE DIAMOND MINES OF SOUTH AERICA 




WINNING THE DIAMONDS 




364 THE DIAMOND MINES OF SOUTH AFRICA 




De Beers Floors. 

old unit of measurement, 16 cubic feet, has been retained, and 
the automatic counters are so geared that every time four 
20-cubic-feet trucks pass them, five truck-loads are registered. 
Each of the rectangular sections of the De Beers floors holds 
about 50,000 loads. The Kimberley floors are nearly as large, 
and substantially the same method is employed in covering 
them. On the depositing ground a truck-load is spread out to 

I '- 




Mechanical Haulage, Kimberley Mine. 



WINNING THE DIAMONDS 365 

cover about 2 1 square feet. So over the miles of floor surface is 
outstretched an enormous carpet of " blue " somewhat less than 
a foot in thickness, and sprinkled with invisible diamonds. It 
may appear to the reader that the word "invisible" is used to 
convey the idea that the diamonds are very small, but such is 
not the case, for many of the diamonds lying buried are as big 
as filberts, and it is not unusual to find them as large as walnuts. 
What is meant is that the diamonds contained in the blue 
ground are invisible to one walking casually over the floors even 




De Beers Floors. 

after the ground has pulverized. During the fifteen years of 
my charge of De Beers mines I have never found a diamond on 
the floors. 

It will be seen that no pains have been spared to hasten and 
cheapen the flow of ground to the floors. After the blue 
ground has been spread out, it is necessary to wait patiently 
until the sun and the rain have contributed their service in dis- 
integrating the breccia. The effect of the exposure of this curi- 
ous compound to heat and moisture is very remarkable. Large 
pieces of blue, which are as hard as sandstone when freshly taken 
from the mine, soon begin to crumble on the depositing floors. 
To hasten the disintegration, the bed of blue is harrowed several 



366 THE DIAMOND MINES OF SOUTH AFRICA 

times to turn up the bigger lumps and expose fresh faces of 
the ground to the sun. Spans of mules were originally used to 




De Beers Floors. 



drag the light harrows used in those days, but steam traction 
engines are now employed to draw wheeled harrows with huge 




Kimberley Floors. 



teeth to and fro across the floors. So the great spread of the 
floors looks like some vast ploughed farm where the laborers 
are preparing the soil for seed. 



WINNING THE DIAMONDS 



367 



The length of time required to effect the desired degree of 
pulverization depends on the season of the year and the amount 




Harrowing Blue Ground with Steam Traction Engines. 

of rainfall. It is curious to note, also, that there is a marked 
difference in the rapidity of disintegration of the blue ground in 
each of the four mines. The blue from Kimberley mine 
becomes well pulverized in three months with heavy rains in 
the summer season, while the De Beers blue requires double 
that time. The longer the ex- 
posure, the more complete the 
pulverization, and the better 
for washing. The long con- 
tinuance of droughts, which 
are of frequent occurrence, 
causes very costly delay. 
During a period of more 
than eight months in 1897 
there was not sufficient rain 
to wet the blue ground. 

_. . r . Harrowing Blue Ground. 

The lack or rain water was 

offset, in a measure, by artificial means ; but as the blue ground 

upon De Beers and Kimberley floors covers 2000 acres of land, 




368 THE DIAMOND MINES OF SOUTH AFRICA 

the difficulty of any approach to complete watering may be 
readily imagined. Under normal conditions soft blue ground 
becomes sufficiently pulverized in from four to six months, but 
it is better to expose it for a longer period, even for a whole year. 




Traction Engine for Harrowing Blue Ground. 

A certain percentage of the blue ground is not affected by 
exposure on the floors. This intractable ground, which is called 
hard blue, makes up about 5 per cent of the product of De 
Beers mine. The large pieces of hard blue are removed from 
the floors to be crushed in rock breakers and rolls, and large, 
worthless boulders and stones embedded in the blue, as well as 
large pieces of basalt and shale which fill the open mines, and 



WINNING THE DIAMONDS 



3 6 9 



have become mixed with the blue ground during the process 
of mining, are picked out to be thrown away. Then the well- 




disintegrated blue ground is taken from the floors in trucks by 
endless rope haulages to the washing machines and put through 
the first stage of concentration. 

The ground is dumped from the trucks into hoppers, at 
the bottom of which are small revolving tables upon which the 
ground is divided and fed automatically into two revolving 
cylinders. This automatic feeder, which was devised by Mr. 
Robeson, late mechanical engineer to De Beers Company, not 




Loading Pulverized Blue Ground. 



only divides the ground equally between two rotating washing 
machines, but delivers it regularly, so that the machines can- 
not become overcharged, which would result in loss of diamonds. 



2 B 



370 THE DIAMOND MINES OF SOUTH AFRICA 




Washing Machine, De Beers Mine. 

After leaving the automatic feeders, the ground is mixed with 
puddle (the name applied to the thick muddy water which flows 
out of the washing pans) and a quantity of clear water is added. 




No. 2 Washing Plant, De Beers Floors. 



WINNING THE DIAMONDS 



This mixture serves to bring the fresh supply of blue ground in 
the pans to the proper consistency for washing, for experience 
proves that diamonds and the heavy minerals with them separate 
from the mass of lighter material much better in a fairly thick 
puddle than in comparatively clear water. From the chutes 
below the feeders the mixture flows into a revolving cylinder 




covered with perforated steel plates with holes i^ inches in 
diameter. All lumps larger than the holes pass out of the end 
of the cylinder, and are carried by a pan conveyor to crushing 
rolls for further treatment. Worthless stones carried in the 
ground are picked out by hand as the lumps move along on the 
conveyor. 

The pulverized ground which passes through the screen holes 
of the cylinders is fed into shallow circular pans, divided so as to 
form an annular space, four feet in diameter, between the outer 
and the inner rim (see figures on pages 372-373). Here the 
ground is swept around by revolving arms attached to a vertical 
shaft, and carrying wedge-shaped teeth (see figure). These teeth 
are set to form a spiral which forces the diamonds and other 
heavy minerals to the outer side of the pan, while the lighter 



372 THE DIAMOND MINES OF SOUTH AFRICA 

material flows out of the discharge situated upon the inner 
rim. Fifty per cent of De Beers ground, when well pulver- 





DETAILS OF ROTARY 



ized, will pass through a screen with holes -^ of an inch 
square, and 66 per cent of Kimberley ground will pass 
through the same screen. The big pieces of hard rock, which 
were brought out of the mines only a few months before, have 
crumbled almost to dust, which, during every working day in the 
year, passes through the pans in a flowing stream for ten hours a 



WINNING THE DIAMONDS 



373 



day, leaving its treasure behind. When the bare statement is 
made that nearly five million truck-loads, or more than four mill- 
ion tons of blue ground, have been washed in a year, the mind 
only faintly conceives the prodigious size of the mass that is 
annually drawn from the old craters and laboriously washed and 
sorted for the sake of a few bucketfuls of diamonds. It would 
form a cube of more than 430 feet, or a block larger than any 
cathedral in the world, and overtopping the spire of St. Paul's, 




WASHING MACHINE. 



while a box with sides measuring two feet nine inches would 
hold the gems. 

When the day's work is completed, the pans, through each 
of which three hundred loads have passed, are emptied or 
" cleaned up," and the concentrated deposits of diamonds, 
mingled with the other heavy but valueless minerals, are then 



374 THE DIAMOND MINES OF SOUTH AFRICA 

sent to the Pulsator in trucks with locked covers, where they are 
sized by passing through a cylinder covered with steel sieving with 
holes from one-sixteenth to five-eighths of an inch in diameter. 
The five sizes which pass through the cylinder flow upon a com- 
bination of jigs, termed at the mines the pulsators, and the name 
Pulsator, which originally applied to the one set of jigs only that 
did all the work for the De Beers Mining Company in 1886, is 
still applied to the large concentrating plant and machinery where 
the final concentration is done and the diamonds sorted from the 
worthless minerals with which they are associated. 




De beers Crushing Mill, Back View. 

Before tracing the diamonds through the Pulsator, it is desir- 
able for the sake of clearness to sketch the treatment of the hard 
blue ground taken direct from the depositing floors. For the 
handling of this portion of the product of the mines an elaborate 
and costly plant was erected on one of the old tailing heaps. 
The driving power of the crushing mill is a compound vertical 
engine of 1 100 horse-power. The whole plant is divided into four 
sections, and provided with friction clutches so that any portion 
of the machinery may be stopped without interfering with the 
running of the rest of the mill. 



WINNING THE DIAMONDS 375 

An endless wire rope haulage carries all the refractory ground 
to the mill, where it is put through a series of crushing machin- 
ery. The first or " comet " crushers reduce the ground so that 
the largest pieces will pass through a two-inch ring. From these 
crushers the ground passes through revolving screens which sep- 
arate the finely crushed from the coarse pieces. The fine size is 
conveyed to the washing pan, and the coarser ground passes from 
the end of the screen to revolving picking tables, where diamonds 
of the larger size may be seen and removed without risk of 
crushing by further pulverization. From the picking tables the 
ground is scraped automatically into two sets of rolls, and the 
pulverized product screened again and graded into three sizes. 
The finest size, passing a half-inch screen, goes to the washing 
pans, and the two coarser sizes to jigs. Large diamonds which 
have been separated from their envelope of blue are retained in 
the jig. The ground still holding the smaller diamonds passes 
out of the end of the jig and then through a series of rolls, 
screens, and jigs until the finished product is drawn from the bot- 
tom jigs into locked trucks running on tramways to the pulsator 
for further concentration and sorting. 

From beginning to end of this process the crushed ground 
is carried by water, and the plant requires a flow of 400,000 gal- 
lons an hour. After leaving the last jig the water is separated 
from the fine ground by a revolving screen and the tailings are 
taken away in trucks to the tailing heap. Within the past three 
years the ordinary rotary pans have supplanted the jigs, and are 
found to be more economical. 

The coarse ground, which passes out of the end of the revolv- 
ing cylinders of the washing plants, is called " lumps." As the 
lumps leave the end of the cylinders they fall upon a conveyor 
and are taken to the end of the washing machines, where they 
are reduced by a similar, though smaller, crushing plant, with the 
exception that pans only are used for saving the diamonds. 

Thus the screened and sized product from the washing pans 
and the crushing machines reaches the final stage of concentra- 
tion in the Pulsator. This is a combination of jigs with station- 



376 THE DIAMOND MINES OF SOUTH AFRICA 




The Pulsator. 



ary bottoms covered with screens with square meshes. The 
meshes are a little coarser than the perforated plates of the 
cylinders that size the concentrate for the jigs. Upon the jig 
screens a layer of leaden bullets for the finer sizes and iron for 
the coarser sizes is spread, forming a bed that prevents the 
deposit from passing through the screen too rapidly. The 
heaviest part of the deposit with the diamonds passes through 
the screens into pointed boxes, from which the deposit is drawn 
off and taken to the sorting tables. The lighter material or 
refuse flows over the ends of the jigs into trucks, which are 
hauled away and dumped on the tailing heap. 

Only one per cent of the total amount of ground washed, or 
one in a hundred loads, goes to the Pulsator in the form of con- 
centrate. Eight and a half per cent of this passes through the 
screens below the five-eighth inch size, thirty-three and a half 
per cent is above that size, and the balance, fifty-eight per cent, 
flows over the jigs as waste. Formerly, for every hundred loads 
washed, five-twelfths of a load passed over the sorting tables, 



WINNING THE DIAMONDS 377 

ordinary wooden tables covered with steel plates. Here the dia- 
monds were picked out by hand, first by white men while the 
deposit was wet, and later, when dry, by native convicts. The 
concentrate was worked over as long as the cost of handling was 
repaid by the gleaning of diamonds. The size of the stones 
which reached the sorting tables ranged from one-sixteenth of an 
inch to one and one-eighth inches. 




Sorting Gravel for Diamonds. 

Mixed with the diamonds in the concentrates are a number 
of other minerals of high specific gravity, and some of notable 
beauty though they have no marketable value. Among these 
are the rich red pyrope, the flesh-colored zircon, the blue 
disthene, bright green chrome diopside, pale green rhombic 
pyroxene, and olivine occasionally in large, polished pebbles. 
Some of the garnets are of fine quality, and one was recently 
cut which resembled a pigeon-blood ruby, and attracted an 
offer of ^25. The complete list of minerals found on the 
sorting tables includes: (i) pyrope, having a specific gravity 



378 THE DIAMOND MINES OF SOUTH AFRICA 

of 3.7 and containing from 1.4 to 3 per cent of oxide of 
chrome; (2) zircon (specific gravity 4.41 to 4.7), in flesh-colored 
grains and fragments, but no crystals this mineral is com- 
monly known on the Diamond Fields as Dutch boart ; (3) dis- 
thene, or cyanite (specific gravity 3.45 to 3.7), discernible by its 
blue color and perfect cleavage ; (4) chrome diopside (specific 
gravity 3.25 to 3.5), in fragments bright green in color and con- 
taining, according to Knopp, over two per cent oxide of chrome ; 
(5) enstatite or bronzite with pale green rhombic pyroxene (spe- 
cific gravity 3.1 to 3.3); (6) mica (specific gravity 2.7 to 3.1); 
(7) magnetite (specific gravity 4.49 to 5.2), occasionally found in 
octahedron crystals; (8) non-magnetic iron ore (specific gravity 4.5) 
containing chrome and titanium in varying quantities ; that is to 
say, sometimes it is chrome iron, and sometimes titanium iron 
ore : according to analysis by Knopp, it contains from 13 to 61 
per cent of oxide of chrome and from 3 to 68 per cent of titanic 
acid; (9) hornblende (specific gravity 2.9 to 3.4); (10) barite 
(specific gravity 4.29 to 4.3) ; (n) calcite (specific gravity 2.7) ; 
(12) pyrite (specific gravity 4.83 to 5.2); (13) olivine (specific 
gravity 3.3). 

The work of picking out the diamonds by hand from the 
concentrate on the sorting tables was, of course, necessarily slow 
and tedious. It was the only division of diamond mining and 
winning which seemed beyond the application of blind and 
unconscious machinery. But men to-day are not inclined to 
admit that anything greatly worth doing is impossible. 

A series of experiments was initiated by me with the object 
of separating the diamonds from the heavy valueless concen- 
trates with which they are associated. An ordinary shaking or 
percussion table was constructed, and every known means of 
separation was tried without success. One of the employes of 
De Beers, Mr. Fred Kirsten, was in charge of the experimenting, 
under the supervision of the late Mr. George Labram, the man- 
ager of the large crushing plant, and afterward mechanical engi- 
neer to the company. Notwithstanding the fact that the specific 
gravity of the diamond (3.52) was less than that of several of 



WINNING THE DIAMONDS 



379 



the minerals associated with it, so that its separation would seem 
a simple matter, it was found in practice to be impossible owing 
to the slippery nature of the diamond. The heavy concentrates 
carried diamonds, and diamonds flowed away from the percus- 
sion table with the tailings. When it seemed that every resource 
to do away with hand sorting had been exhausted, Kirsten asked 
to be allowed to try to catch the diamonds by placing a coat of 
thick grease on the surface of the percussion table with which the 




Automatic Diamond Sorter, called the Greaser. 

other experiments had been made. Kirsten had noticed that oily 
substances, such as axle grease and white or red lead, adhered to 
diamonds when they chanced to come into contact, and he argued 
to himself, if these substances adhered to diamonds and not to the 
other minerals in the concentrates, why should not diamonds adhere 
to grease on the table and the other minerals flow away ? In 
this way the remarkable discovery was made that diamonds alone 
of all minerals contained in the blue ground will adhere to grease, 
and that all others will flow away as tailings over the end of the 
percussion- table with the water. After this was determined by 



380 THE DIAMOND MINES OF SOUTH AFRICA 

thorough experiments, more suitable shaking tables were con- 
structed at the Company's workshops. These were from time 
to time improved upon, until now all the sorting (except for the 
very coarse size) is done by these machines, whose power of dis- 
tinction is far superior to the keenest eye of the native. Since 
the discovery of the affinity of grease for diamonds, experiments 
have been made with rubies and sapphires from Burma, and it 
was found that grease caught these gems with the same certainty 
that it catches diamonds. 

After a thorough trial a number of these unique diamond- 
catching tables (see cut, p. 379) were constructed, and are now 
working on De Beers concentrates. Each shaking table is 
made of corrugated cast-iron plates in five sections, with a drop 
of about an inch from one division to another. Thick grease is 
spread on the plates to cover them to the top of the corrugations. 

The concentrates are conveyed from the jigs upon a con- 
veyor belt and deposited into hoppers, where the load is elevated 
to revolving cylinders covered with perforated steel plates. 
Through the graded screens of these cylinders the concentrates 
pass into small hoppers, one above each table, fitted with auto- 
matic feeders, cast-iron cylinders with grooves corresponding 
to the graded sizes of the concentrates, and are distributed 
evenly across the upper portion of the shaking tables, and car- 
ried down by a flow of water from a trough fixed behind the 
feeders. During the time the table is working it is rapidly 
shaken from side to side by an eccentric placed on a shaft under 
the table. 

Strange to relate, the descending diamonds stick on the face 
of the grease while all other minerals pass over it. Only about 
one-third of one per cent of diamonds is lost by the first table, 
and these are recovered almost to a stone when the concentrates 
are passed over the second table. The discrimination of this 
sorter is surely marvellous. Native workers, although experi- 
enced in the handling of diamonds, often pick out small crystals 
of zircon, or Dutch boart, by mistake, but the senseless machine 
is practically unerring. It will catch rubies, sapphires, and emer- 



WINNING THE DIAMONDS 381 

aids as well as diamonds, but so far as it has been tested, it will 
not cling to anything but a precious stone. The grease which 
is used loses its power to catch diamonds after a few hours' work, 
owing to its becoming more or less mixed with particles of 
water. It is then scraped off the tables, together with the dia- 
monds adhering to it, placed in a kettle made of finely perfor- 
ated steel plates, and steamed. The grease passes away to tanks 
of water, where it is cooled and is again fit for use. The dia- 
monds, together with small bits of iron pyrites, brass nails from 




The Manager of the Pulsator, Mr. James Stewart, through whose Hands ^3,000,000 to 
,4,000,000 Worth of Diamonds pass every Year. 

the miners' boots, pieces of copper from the detonator used in 
blasting, which remain on the tables owing to their high specific 
gravity, and a very small admixture of worthless deposit which 
has become mechanically mixed with the grease, are then boiled 
in a solution containing caustic soda, where they are freed from 
all grease. The quantity of deposit, from the size of five-eighths 
of an inch downwards, which now reaches the sorting table, does 
not exceed one cubic foot for every 12,000 loads (192,000 cubic 
feet) of blue ground washed. As already stated, one-twelfth of 
one per cent of the whole mass of blue formerly passed to the 



382 THE DIAMOND MINES Of SOUTH AFRICA 




WINNING THE DIAMONDS 383 

sorting tables; or, from 12,000 loads, which is about the daily 
average of the quantity washed at De Beers and Kimberley 
mines, 160 cubic feet had to be assorted by hand. 

The first question usually asked by visitors is, What is the 
cause of this amazing discrimination ? This is a very difficult 
question to answer with positive assurance. It is possible that 
the secret of the affinity may lie in the fact that water adheres 
to or enters into all minerals composing the concentrate except 
precious stones. These present comparatively dry faces to the 
grease and quickly adhere to it, while the wet stones flow over 
the table. The grease has no affinity for a piece of glass, which, 
when dropped on the table, flows away in the tailings. 

From the sorting tables the diamonds are taken daily to the 
general office under an armed escort and delivered to the valua- 
tors in charge of the diamond department. These experts clean 
the diamonds of any extraneous matter, such as small particles 
of adhering blue ground, by boiling them in a mixture of nitric 
and hydrochloric acids (aqua regia), or, still better, in fluoric 
acid. When the stones are cleaned, they are carefully assorted 
with reference to size, color, and purity, and made up in parcels 
for sale, formerly to local buyers, who represented the leading 
diamond merchants of the world. For several years past De 
Beers Company has sold in advance its annual production to a 
syndicate of London diamond merchants who have representa- 
tives residing in Kimberley. 




A Day's Diamond Wash. 



CHAPTER XIII 



OBSTACLES AND PERILS 




N the open workings the imminent hazard of 
maiming and death by reef slides was ever 
hanging over the heads of the miners. In 
view of the rashness with which the pit sink- 
ing was pressed, it was a marvel, indeed, that 
the actual loss of life was, on the whole, so 
small. No complete or accurate records were ever kept of the 
men injured or killed in prosecuting the work before the advent 
of systematic mining. 

In the journals of the Diamond Fields the most noteworthy 
casualties were recorded, and it is seen that in the years immedi- 
ately following the undertaking of underground mining, the 
principal loss of life occurred from the falls of loosened pieces 
of blue ground or reef. This is expressly noted in the report 
of the Inspector of Mines at Kimberley to the Assistant Com- 
missioner of Crown Lands on August 27, 1885. Underground 
mining operations in Kimberley and De Beers mines were then, 
he observed, becoming very hazardous. In both mines, but 
especially in the Kimberley mine, " some of the underground 
working places in diamantiferous ground are huge caverns of 
from 25 to 52 feet in height and 20 to 30 feet in width. The 
roofs of these workings, from exposure to atmosphere, shocks 
of blasting, and inherent weakness of the blue or diamantiferous 
ground, are becoming extremely unsafe ; occasionally pieces of 
the ground or rock fall from the high roof or sides, to the immi- 
nent danger of persons working on the floors. During the last 
and current months there have been three deaths in under- 
ground working places directly due to the dangerous operations 

384 



OBSTACLES AND PERILS 385 

in the mines," and in view of this danger and loss of life, the 
inspector urgently recommended the limitation by the govern- 
ment of the height and width of the underground workings. 

" Main tunnels to be used only for traffic not to exceed 8 
feet in width and 8 feet in height. 

" Working chambers or stalls from which the blue or dia- 
mantiferous ground is excavated in bulk, not to exceed 18 feet 
in width by 20 feet in height to the highest point. 

" Partitions or pillars not to be of less thickness than half 
the width of the contiguous chambers or stalls. 

"The roof of ceiling between one level and the next above 
to be not less than 20 feet in thickness at the highest point of 
the lower workings." 

This recommendation had in view obviously the precautions 
enforced in the working of coal mines, and would doubtless 
have afforded an increased measure of protection, but the 
method of working proposed was not well suited to the develop- 
ment of the diamond-bearing ground, as was later conclusively 
determined. The slaking and crumbling of the diamond-bear- 
ing breccia upon exposure to air and moisture make roof falls 
and slips from the sides especially frequent and disastrous. 
The ground is full of soapy seams, and pieces of considerable 
size drop without a moment's warning, so that it is necessary, 
in places, to keep the tunnels timbered as near the working face 
as possible. Risk from this cause cannot be wholly obviated in 
such mining, but the introduction of the new system adopted 
for the working of the mines, shortly after they came under my 
management, has greatly diminished this peril, and the resultant 
loss of life or injury to the workmen. By the new system the 
levels are worked back from the surrounding hard rock or reef 
in sections, formerly 30 feet, now 40 feet apart, as before partic- 
ularly described, in a series of terraces, extracting the ground 
from the uppermost level downward in succession. This 
method did away with any danger of collapse in the under- 
ground works, and by successively robbing out the roof and 
sides of the tunnels on each descending terrace, the caving of 
z c 



386 THE DIAMOND MINES OF SOUTH AFRICA 

the unstable ground was systematically anticipated and restricted. 
No feasible care in the direction of men working in such shifting 
ground can entirely do away with casualties. Some are scarcely 
to be avoided, but most are attributable, more or less, to the 
miners' heedless disregard of the warnings of overseers and 
proper precautions. 

There was another serious risk in mining in the upper 
levels of the mines, where shale is heavily impregnated with 
bituminous matter, and no device could wholly prevent the 
gathering of carburetted hydrogen, which, mingling with air, 
forms the " fire damp " that has been so deadly a peril to miners. 
When sinking shafts or driving tunnels in the shale, miners are 
prohibited by the strictest injunction of the management, and 
the formal regulations of the Government Inspector of Mines, 
from carrying any lighted candle into passages where there is 
any possibility of this gas having gathered; but no prohibition 
has ever been able to prevent an occasional stretch of reckless- 
ness on the part of some careless miner. Locked safety lamps 
are provided abundantly for testing the atmosphere in such 
parts of the mine workings, but neglect of this precaution has 
caused startling explosions, scorching and striking men down, 
and in a few cases causing death. In 1883 there was a slight 
explosion of accumulated gas in the reef workings of the 
French Company, Kimberley mine. Here thin bands of coal 
had been struck in the black shale, and in an upward drive 
to meet a pass, some gas had collected in the interval from 
Saturday to Monday. A naked flame set fire to this gas and 
caused the explosion. Prior to this time two other cases were 
on record, in both of which workmen were severely injured. 
Perhaps the most notable instance of the gathering of this gas 
was in a heading of the workings of the Gem Company in De 
Beers mine in July, 1885. One of the workmen had his face 
and hands badly scorched by an explosion at the end of the 
heading, and a second explosion occurred shortly afterward, 
when the managing director and an overseer attempted to exam- 
ine the heading, taking candles to light their way. The director, 



OBSTACLES AND PERILS 387 

Mr. George McFarland, was severely burned by this blast of 
gas, which was described as a "fizz" almost noiseless. Since 
the workings have been carried down below the level of the 
shale, there has been no danger from fire damp, and the acci- 
dents from this cause have ceased to occur in the deeper mines. 

The strictest precautions are enjoined in the storing and 
handling of explosives used in the diamond mines, and the 
need of such stringency was signally emphasized in the destruc- 
tive explosion that wrecked a dozen magazines near the com- 
pound of the Victoria Mining Company on October 31, 1884, 
three years before I took the management of De Beers. The 
shock was felt from Dutoitspan to the farthest limits of the west 
end of the camps, and terror-stricken people rushed out of their 
houses to see a vast heaving cloud of smoke rising hundreds of 
feet into the sky. 

The magazines were dashed to pieces, as the Kimberley 
papers reported, by the terrible power of the explosives. In 
most instances the galvanized iron was broken into tiny atoms 
as if by myriad hammers, and cartridges were scattered far and 
wide through the debris, exploding in volleys or scattering blasts 
for many minutes after the explosion. One large stone was 
thrown as far as the Central Company's offices, a distance of 
two miles, and smaller ones to the West End, three miles from 
the magazines. In the most distant parts of the camp there 
was a startling breakage of windows, lamps, and chandeliers, and 
the hotel bars and canteens were so heavily pelted that " the 
floors were swimming with what we might call dynamite cock- 
tail, composed of every liquor under heaven from Cape Smoke 
to Heidseck and Pommery." Witnesses of the explosion thought 
that hundreds had been killed and injured, but almost miracu- 
lously, as it seemed, only two persons were killed, one a white, 
the other a black, both bodies being horribly mutilated. A 
third sufferer was taken up and tenderly cared for, a poor native 
deeply gashed and with broken ribs. 

No other accidents in the mines have ever approached in 
loss to life the terrible disaster from the outbreak of fire in De 



388 THE DIAMOND MINES OF SOUTH AFRICA 



Beers mine in July, 1888. When the Consolidated Mines took 
over the property of De Beers Mining Company, nearly all the 
blue ground was hoisted from the foo-foot level, through the 
first large working shaft constructed, known as No. i west end 
incline. In July, 1888, another shaft, No. 2 incline, had just 
been completed to the yoo-foot level, and skips in the y-foot 
compartment were used in hoisting the ground broken on this 
level. In addition to these working shafts a small vertical pros- 
pecting winze, called the Friggin's shaft, had been sunk from 
the foo-foot to the yoo-foot level. When a tunnel connec- 
tion was opened between No. i and No. 2 inclines on the yoo- 
foot level, the prospecting winze 
was no longer needed, and it 
stood abandoned except as a 
ladderway. There was a small 
disused engine room on the 500- 
foot level a short distance from 
the winze. With the sinking 
and connection of both working 
shafts on the yoo-foot level, the 
output of the mine increased 
until a total of 104,089 loads 
was attained during the month 
of June, 1888. 

On the 9th July following, 
large skips in No. 2 incline be- 
gan carrying blue ground from 
the yoo-foot level, and continued hauling until the morning of 
the nth, when one of them jumped the rails, either because the 
hoisting was being done at too rapid a pace, or from some 
obstruction in the shaft. Examination showed that both skips 
were off the rails, and that the shaft timbers had been consider- 
ably damaged. In bringing up one of the small skips in the 
manway, this was also derailed by the debris in the shaft. The 
necessary work of repair was begun at once and continued during 
the day. During the changing of the shifts in the evening, the 




Mr. Lindsay, Mine Manager, killed in 
De Beers Mine Fire, July II, 1888. 









OBSTACLES AND PERILS 



389 



mine manager, Mr. Lindsay, reported that the work was pro- 
gressing as fast as practicable and that the shaft would be in run- 
ning order within a few hours. 

At about half-past six o'clock in the evening Lindsay and 
six miners went down the shafts in one of the small skips. 
A few minutes later an alarm of fire was given just as I was 
about to drive to my home from the works. It was reported 
to me through the telephone that the Friggin's shaft was on 
fire. It is probable that one of the native miners had sneaked 




Fire in De Beers Mine, July II, 1888. 

off to the disused engine room on the 5oo-foot level, and placed 
a lighted candle so carelessly that the flame ignited the timbers, 
perhaps while the lazy savage was snoring on the floor. The 
precise cause of the fire was, however, never determined, but 
from the time of its starting, it spread with such swiftness that 
it could not be stifled. 

Within a few minutes after the outbreak of the fire both of 
the incline shafts were filled with dense smoke, as both shafts 
were upcasts, and the passage of any of the men through these 
exits from the mine workings was hopelessly shut off. When 
the alarm was given, there were 685 men at work in the levels 
below the fire, and our anxiety for their safety may be readily 
conceived. At the first warning of danger two men were sent 



390 THE DIAMOND MINES OF SOUTH AFRICA 

down No. 2 incline to notify Lindsay and his companions 
of the outbreak of the fire, but the smoke came up through 
the shafts so heavily that both were driven back gasping for 
breath, and barely reached the surface before they fell on the 
floor completely exhausted. For several minutes there was a 
tension of waiting for some signal to hoist from Lindsay, or one 
of his party, but none was given. Lindsay and his comrades 
must have been close to the skip in the shaft when the fire 
started, and a signal bell wire ran through the shaft close at 
hand. There was time enough for one of the party who went 
down the shaft in the skip with Lindsay to climb up the 
shaft by means of the timbers, a distance of 150 feet, and in 
view of this, the failure of these men to get into the skip and 
ring a signal to hoist is inexplicable. Seeing at once that ascent 
through No. 2 incline was probably hopelessly blocked by the 
outpouring smoke, I hastened to the mouth of the other shaft 
(No. i incline shaft). The smoke was also streaming out of 
this shaft in dense volumes. 

The signal to hoist men by ringing three bells was repeatedly 
given, but I hesitated to give the order to hoist the skip, which 
was at the 6oo-foot level, as the risk of hoisting a skip-load of 
men through the stifling smoke was appalling. On the other 
hand, it was impossible to know at the surface in what desperate 
straits the men might be on the 6oo-foot level. So, before giv- 
ing the signal to hoist, I took measures to revive the men who 
would be overcome by the smoke in ascending the shaft, and 
water was provided to dash on them if they came up with their 
clothes on fire. It was a moment when no balancing of proba- 
bilities could determine the decision. There was a desperate 
chance of safety in the swift pulling up of the skip. I could 
not let the piteous appeals go on apparently unheeded. I gave 
the signal to hoist at top speed in response to the last pleading 
signal. When the skip was about 300 feet from the surface, the 
wire winding rope parted. The broken end came whizzing up 
through the shaft, but the skip with its load of four poor victims 
fell crashing down to the sump at the bottom of the shaft, a 



OBSTACLES AND PERILS 391 

little below the 6oo-foot level. When the rope was examined, it 
was found that the flames from the burning timbers had made 
it so hot that the tension of the skip drew out the wires to fine 
needle points which snapped under the strain. When the first 
signal to hoist was given, there were ten or twelve men in the 
skip, but the majority left it when the signal to hoist met with 
no response. It was impossible for the men at the 6oo-foot 
level to know that the shaft through which they wished to be 
hoisted was on fire a hundred feet above them, nor could we on 
the surface know what was happening 500 feet below. 

The mine was ventilated at the time through an outlet into 
the old open workings, and through the Gem shaft on the east 
side of the mine. The Gem shaft was a small, old working 
shaft that had been sunk from a terrace in the blue ground. 
Unfortunately it had been partially closed by a recent ground 
slide in that part of the mine. It was, however, still sufficiently 
open to be of invaluable ventilating service at this crisis, and it 
could have been opened for the rescue of the men in the mine 
if there had been no other means of escape through the outlet 
into the open workings. During the hours of fearful anxiety 
that followed the closing of the two main shafts, the outlet from 
the mine to the open workings was intently watched, and daring 
parties penetrated far within it in the hope of communicating 
with miners escaping from the range of the fire. Almost all of 
the men in the mine were well acquainted with this passage to 
the surface, and it was confidently hoped that many, at least, 
would contrive to grope their way upward through this outlet 
to safety. Fortunately the air draught through this passage 
was downcast, and the inrush of air cleared the passage from 
smoke. 

To the immeasurable relief of all, so anxiously expectant, 
one white man and six native miners came climbing through 
this passage into the open workings at about ten o'clock on the 
night of the fire. This showed that a practicable way of escape 
from the mines was open, but many hours of fearful suspense 
followed throughout that night and the following day, while the 



392 THE DIAMOND MINES OF SOUTH AFRICA 

miners were groping their way to the surface through the same 
opening. Forty-two white men and 441 native miners were 
thus rescued, but 24 whites and 178 natives lost their lives 
in levels and passageways charged with deadly smoke. The 
downcast draught through the Gem shaft was the salvation of 
the greater part of the rescued men, who spent this fearful night 
on the level close to this shaft, which was free from smoke. 
During the afternoon of the following day, July 12, a party of 
heroic men penetrated far into the mine through the entrance 




No. 2 Incline Shaft, looking East. 

in the open workings, and rescued a number of natives who 
were cowering stupefied by the smoke, or paralyzed by fear. In 
this rescuing party were some who had passed the night in this 
frightful prison, but who were, nevertheless, among the first to 
volunteer to go down again in the desperately hazardous venture 
to save their comrades. 

No. i incline was completely burned out and caved in during 
the night of the fire. During the night of the I2th No. 2 
incline caved in also for a distance of about 40 feet, near the 
junction of the shale with the hard rock, shutting off" all com- 
munication with the mine. Before the latter shaft could be 




.4* 




OBSTACLES AND PERILS 



393 




394 THE DIAMOND MINES OF SOUTH AFRICA 

reopened, the water in the mine rose to a depth of 20 feet, filling 
all the tunnels on the yoo-foot level. 

Several days after the fire I went down the shaft accompanied 
by Captain Hambley, Assistant Inspector of Mines, and one of 
the overmen. I arranged to lower the skip gradually down the 
incline to make the first inspection. As we went down, an 
insulated signal wire was lowered, and provision was made so 
that I could keep the bell ringing continually, and instructions 
were given to haul up the skip at the moment the ringing 
stopped, for I feared that we might drop into foul air so sud- 
denly that we would not be able to signal in the usual manner. 
So we went down in the skip slowly to a point about 150 feet 
above the crushed ground in the shaft. At this point, some 250 
feet below the surface, we saw the body of one of the men who 
went down with Mr. Lindsay just before the breaking out of 
the fire. We did not stop, for the moment, but kept on signalling 
until the skip was lowered to the ground which closed the shaft. 
Our search for any further trace of the lost miners was fruitless, 
for we could find no more bodies. Mr. Lindsay and his remain- 
ing companions were buried beneath the debris when this part of 
the shaft caved in. Finding that the further descent of the skip 
was cut off, I then gave the signal to hoist, and on reaching the 
surface, gave instructions for men to go down and remove 
the body seen in the shaft. The poor man had climbed up to 
the point where he died, in a desperate effort to escape. The 
other men, as well as the skip in which they went down, were 
buried deeply under the mass of crushed ground. 

The work of repairing No. 2 incline could not be begun 
until July I9th, for the smoke and heat from the mine made 
work in the crushed portion of the shaft unendurable. Even 
then it was only practicable to advance very slowly, and the shaft 
was not opened until the jd of August, when the large skips 
were at once employed to bail out the water. Eight days later 
the mine was drained, and the reopening of the workings could 
be undertaken. 

It was originally intended that the large skips in No. 2 in- 



OBSTACLES AND PERILS 



395 




396 THE DIAMOND MINES OF SOUTH AFRICA 

cline should be used in hauling blue ground from the yoo-foot 
level only, as there were ample facilities in No. i and in the 7-foot 
compartment of No. 2 for hoisting all the blue ground taken 
from the 6oo-foot level and the levels above. Consequently no 
stations had been made ready for the larger skips on the latter level. 
It was necessary, therefore, to open tunnels, sink passes, and put 
in chutes to connect the 6oo-foot level with the surface, besides 
excavating a pump chamber and erecting new pumps, before the 
regular output of blue ground could be resumed. During the 
month of August only 8613 loads were hauled, and this was 
mostly of poor quality from excavations of the west end of the 
mine. During September the output was increased to 57,408 
loads, in October to 87,225 ; but it was not until the following 
month of November that the output reached 104,285 loads, or 
approximately the same amount as in the month before the fire. 

This brief sketch may serve to show to the general reader 
something of the terror, the peril, and the disaster which an out- 
break of fire in any great mine may cause. As soon as practica- 
ble after this fire, the previously designed systematic and thorough 
opening of the mine was advanced. In addition to No. 2 incline, 
the rock shaft (elsewhere particularly described) was completed 
and connected with the mine by a tunnel on the 8oo-foot level. A 
vertical escape shaft was sunk from one of the terraces in the 
open mine to the yoo-foot level. It had a ladderway and a 
single cage compartment, and was connected with seven levels in 
the mine. The Oriental shaft, situated on the east side of the 
mine, was connected with it at the 5Oo-foot level, from which 
all parts of the mines were reached by ladderways. This shaft 
serves to ventilate the mine, and as an important passage for 
escape in case of need. Besides these four shafts there was a 
tunnel into the open mine, which was connected with the lower 
workings by a double ladderway. The Oriental shaft and No. 2 
incline were upcasts. The rock shaft, escape shaft, and the 
tunnel into the open mine were downcasts. 

The first consideration in working a mine is to have a safe 
exit for the workmen, in case a fire breaks out or the mine 



OBSTACLES AND PERILS 397 

becomes flooded by suddenly tapping a large quantity of water, 
and at the diamond mines this precaution is strictly carried out. 
In the early days of underground mining, when many of the 
levels had exits into the open mine, it was necessary, in pro- 
viding numerous escapes for the workmen, to guard against sly 
sallies of natives when there was no danger, because they could 
leave the mine with stolen diamonds, or could go out for the 
purpose of obtaining intoxicating drink, and bring back bottles 
of Cape brandy, called " Cape Smoke," into the mines with 
them. Although the numerous escapes from the mine were 
guarded by watchmen, the dusky Kafirs would come, at times, 
in squads, and overpower the guards and make their escape. An 
ingenious device was invented by our electrician, Mr. Drum- 
mond, by placing a small copper rod directly above the iron 
rungs of the ladders, and connecting both with a battery. Then 
when a man placed his hand or foot upon the copper rod, it bent 
down, completing the circuit, and rung an alarm bell in the 
mine and on the surface. The natives could never quite under- 
stand why they were always met by a posse of white guards at 
the particular place where they were trying to escape. 

In later years, since the mines became deeper, all workmen 
are taken in and out of the mines by means of cages. There 
are double ladderways in the shafts which may be used in case 
of emergency, but there is always a sufficient number of white 
men employed about the tops of these shafts to prevent the 
escape of natives. 

In view of the responsibility resting upon me from my 
acceptance of the General Managership of De Beers Mines in 
the year preceding this great disaster, and the common duty of 
all connected with the mines to do everything practicable to 
save life, to prevent the outbreak of fire, and to guard against 
all contingencies, it is proper to note the warmly appreciative 
recognition accorded by the presiding chairman, Mr. Barnett 
Isaacs Barnato, at the adjourned first annual meeting of the 
shareholders of the De Beers Consolidated Mines Limited. 
Mr. Barnato said in his address to the shareholders : 



398 THE DIAMOND MINES OF SOUTH AFRICA 




OBSTACLES AND PERILS 399 

" I suppose you all remember about the sad calamity by 
which so many poor fellows lost their lives. At this point I 
feel I must pay a tribute of respect to the brave men who 
worked and risked their lives on behalf of those poor fellows 
who perished in the disaster. I remember on that sad occasion, 
which will never be effaced from my memory, and from the 
memories of many who lived in Kimberley at the time I 
remember seeing our respected and able general manager, Mr. 
Gardner Williams, a gentleman to whom I believe no person 
can attach the least blame, working night and day, and doing all 
he possibly could for the relief of the sufferers. That calamity 
was an act of God, or at least we must conclude so, for on the 
very day of the disaster there was an accident in No. 2 shaft, 
which blocked it up to some extent, and the Gem escape shaft 
gave way only a week previously. I, therefore, think that 
calamity was an act of God, and I hope a similar disaster will 
never again be witnessed in Kimberley or elsewhere. In paying 
a tribute of respect to Mr. Williams, who worked all through 
the night when the fire broke out, and to the brave men who 
went into the mine, to try and save their fellows, we must not 
forget that those men risked their lives, that they went down 
into the mine, when millions and millions of loads of reef were 
hanging over them, to open up the shaft so that the men might 
escape. And the result of their work, we know, was that out of 
about seven hundred men in the mine, five hundred escaped. 
Therefore, in passing this tribute of respect to Mr. Williams 
and the men, I feel sure that it will be universally indorsed by 
the shareholders. [Applause.] No more need be said about 
this matter, except that the state of the mine after the calamity 
necessitated a considerable expenditure of money. I think it 
took us three months to get the mine in proper working order, 
and we lost three months' labor, at a cost of something like 
,250,000. The balance sheet only shows about ^30,000, but 
by the loss of blue, etc., the loss to the company was, as I have 
stated, not much short of a quarter of a million of money." 

Providentially, and by the exercise of every feasible precaution, 



400 THE DIAMOND MINES OF SOUTH AFRICA 

there has been no serious spread of fire in the mines since the 
occurrence of this great disaster. 

The chief peril to life and damage to the workings of the 
mine, for a number of years, has come from the destructive 
" mud rushes," as the miners call them. There is no water in 
the blue ground or the mine itself, but the water flowing into 
the mine from the surrounding reef made a muddy mixture 
of the disintegrated shales, decomposed basalt, floating reef, and 
low grade blue ground, which had fallen into the worked out 
section of the open mines. At times the tremendous pressure 
of the shifting ground above forces this mud in vast quantities 
into the working levels of the mine, and the miners do not have 
time to escape this inrushing mass even by instant flight. On 
several occasions tunnels in the mine have been filled to the 
extent of thousands of feet by these rushes in a few minutes. 
As the work in the mines reached the deeper levels, these rushes 
became so frequent that the working of the mines was seriously 
interfered with, and no watchfulness could avert the loss of life. 

In June, 1897, one of the worst mud rushes known in the 
record of the mines occurred in De Beers mine, filling up almost 
instantly a large number of tunnels on the looo-foot level. Two 
native miners were overtaken by the rush, and shut up in a 
drainage passage that was in progress to tap the water in that 
section of the mine. For a stretch of 28 hours they were held 
fast in this narrow prison chamber, momentarily dreading a fur- 
ther rise of the mud that would bury them alive. Meanwhile 
the most daring efforts were made to rescue them from their 
stifling prison, and two heroic men, Thomas Brand and John 
Brown, finally burrowed through 200 feet on the top of the mud, 
and brought the two natives out safely at an appalling risk to 
their own lives. The rescue was barely in time, for the next 
morning another rush followed, filling up the tunnels again still 
further, and rising to the top of the passage that had given 
breathing room to the imprisoned men. For this signal heroism 
medals of the Royal Humane Society were very fitly given to 
Brand and Brown. 



OBSTACLES AND PERILS 



401 




4-02 THE DIAMOND MINES OF SOUTH AFRICA 

In May, 1898, there was another great mud rush through 
the ii2o-foot level, from which a whole gang of native workers 
barely escaped alive. On this occasion "Jim," one of the best 
of the "baas" boys, was almost buried alive with his gang of 15 
men. The rush shut this working party up in a narrow passage 
on this level for more than 64 hours. When the men were 
rescued at length from their stifling quarters, where they were 
imprisoned for more than two and a half days, without a morsel 
of food to eat or a drop of water to drink, all were greatly 
exhausted, as might be supposed. But in spite of his sufferings, 
the brave leader, Jim, went back at once into the mine to 
grope back over the mud in search of one of his gang whom he 
supposed was missing, and he would not return to the surface 
until he learned beyond doubt that all had been rescued. 

The endurance of the native miners under such circumstances 
is remarkable. In July, 1898, a Basuto boy, "Joseph," was 
almost buried in a mud rush, and was completely shut in the 
" dead end " of a tunnel, on the 96o-foot level. The attempt 
to clear a passage to rescue him was begun at once, and the work 
was pushed without a respite night and day, but it was late on 
the third day before the place of his entombment was reached. 
He was found lying crouched beneath some timbers resting on 
an overturned truck, around which the mud had risen to the 
depth of two and a half feet. The rescue party had given up 
all hope of finding him alive, and were about to blast the envel- 
oping mud in order to pull out the truck, when a faint cough 
was heard, apparently coming out of the dense mass of mud. 
The natives at work were badly frightened at this weird sound, 
and called up the contractor in charge, who finally succeeded in 
digging out the poor Basuto boy nearly lifeless. One of his legs 
had been pinned beneath the truck so heavily that the circulation 
of the blood was stopped, and mortification set in, necessitating 
its amputation. The boy bore the operation with the charac- 
teristic fortitude of his race, and is stumping about to-day with 
a wooden leg. He had been shut up for more than three days 
in a little hole in the ground wholly without food and drink, 



OBSTACLES AND PERILS 



403 



and with only a few cubic feet of compressed stagnant air to 
breathe. 

When a tunnel is being driven there is only one way of 
escape, and the working face is called a dead end, though not 
on account of its deadly nature in cases of a mud rush, for 
it is a common term in miners' parlance. In point of fact 
these dead ends are the safest places in the vicinity of a 
mud rush. The mud, which first fills the mouth of the tun- 
nel, forces the air ahead of it, and compresses it to such an 
extent that it checks the advance of the mud. Hence, if a 
native is hemmed in, he has sufficient air to breathe until 
he can be rescued. On more than one occasion when natives 
have been caught in the rush of mud, their narrow cell would 
not have held sufficient air to keep them alive had it not 
been that a large quantity of air was compressed into the small 
space. 

On one occasion two natives were shut up in the dead end 
of a tunnel for ninety-five hours. They had no food, but man- 
aged to obtain a small quantity of water as it trickled down from 
the roof and sides of the tunnel after finding its way through the 
blue ground from the level above. These men had more air 
space than is usually the case, and the temperature in the ends 
of the tunnels ordinarily ranges from 75 to 90 degrees. When 
rescued they were greatly exhausted, but after a few days of 
medical treatment they were quite fit again, and resumed their 
work in the mine. At another time, when natives were shut in 
for nearly two days, they swallowed small balls of soft mud, and 
when rescued it took a considerable time to bring their diges- 
tive organs back to their normal condition. On several occa- 
sions the white miners have been victims to similar experiences, 
and now and again a white miner has lost his life by being 
overtaken and enveloped in the mud. The longest period of 
time that a white man has been confined in the end of a tunnel 
is about two days, and there were a dozen or more natives with 
him. By giving the usual miners' signal of tap-tap tap-tap- 
tap, on the walls of the tunnel, we knew he was alive, and it 



404 THE DIAMOND MINES OF SOUTH AFRICA 

may be imagined that no time was lost in extricating him and 
his men from their perilous position. 

Of recent years the measures described in the preceding 
chapter have proved effective in freeing Kimberley mine from 
this peril. The water which finds its way into De Beers mine 
has not yet been entirely taken up, but by driving tunnels around 
the mine to tap the water the danger has been minimized. On 
the ist of October, 1899, s ' x nat ives were overcome by a mud 
rush and killed. Wherever there is the least sign of mud, the 
workmen are withdrawn, and the places fenced off until the mud 
has come out or the water is drained off, leaving the places safe 
for the miners to reenter them. 

As there have been from ten to twelve thousand men employed 
in the mines and workshops and on the depositing floors, three- 
fifths of whom are underground workers, who are to a greater 
or less extent raw and untrained natives, the percentage of deaths 
and injuries has not been excessive. 

In the painstaking and valuable reports of Dr. C. Le Neve 
Foster, H. M. Inspector of Mines, he compares the returns of 
casualties in the South African mines with the like statistics of 
mines in which trained Englishmen are employed. This com- 
parison bears hardly in its application to the diamond mines, in 
view of the fact that the great majority of the native workers in 
these mines are "raw hands." There is probably a change of 
half the workers in the mine every year, and the men coming 
in to offset the outflow are mostly natives who have not worked 
in the mines, and are familiarly known as "green hands." In 
time these men are trained to a fair measure of proficiency, but 
it is to be expected that the proportion of accidents to the 
numbers of such workmen will be greater than the average in 
English mines. 

From the carefully prepared statistics of Sir Frederic Augus- 
tus Abel, covering the loss of life in English mines, 1 it appears 
that the greatest loss occurs from falls of the roof and sides 

1 Supplement to Forty-fifth Annual Report of the Registrar General of Births, 
Deaths, and Marriages in Great Britain. 



OBSTACLES AND PERILS 405 

of mine workings, amounting to 40.77 of the total. The loss 
of life from explosions comes next, with a showing of 23.17 per 
cent. In the records of fires in mines from all causes, it is 
shown that only a very small percentage of men are actually 
burned to death, fully 90 per cent of the deaths resulting from 
suffocation. 

Contrary to the popular impression, it has been shown by 
Dr. C. Le Neve Foster, that the ore miner has nearly as danger- 
ous an occupation as the coal miner; and in Cornwall and some 
other metalliferous districts the average losses from accidents 
were higher than in coal mines. Dr. Ogle has pushed this com- 
parison farther by his statistical demonstration that, in spite of 
accidents, the death rate of coal miners is not high. In com- 
parative mortality these miners ranked only thirtieth in a list 
of ninety-four occupations ; but the mining in Cornwall, at 
the time of this report, was exceptionally perilous, standing 
ninety-first on the list. In other words, only three of the ninety- 
four occupations exceeded the mining in this district in deadli- 
ness. This peculiarly high mortality was ascribed to inadequate 
ventilation and excessive climbing of ladders from deep mines. 1 
These conditions, of late years, have been bettered. 

1 Supplement to Forty-fifth Annual Report of the Registrar General of Births, 
Deaths, and Marriages in Great Britain. 



406 THE DIAMOND MINES OF SOUTH AFRICA 




CHAPTER XIV 




THE WORKERS IN THE MINES 

OWHERE else on the face of the earth is there 
an assemblage of workers of such varied types 
of race, nationality, and coloring as are to be 
seen in the South African Diamond Fields. 
There is hardly a nation of Europe or Colony 
of the British Empire that has not some repre- 
sentatives. There are adventurers from the United States, 
Mexico, and South America; and white men from all the Colo- 
nies of South Africa mingle with the masses of native Africans 
of every shade of dusky hue shown by the tribes that range 
from the Cape to the equator. Even the American Indian is 
not unknown in the fields, one specimen at least having resided 
there for many years. Add to this motley throng a sprinkling 
of dark East Indians, Malays, and Chinese, and the kaleido- 
scopic shifts and coloring of this babel in the 
Diamond Fields may be dimly conceived. 

Only about a sixth of the workers in the 
mines are whites, and the larger part of these 
are employed above ground on the floors, in 
the workshops, and in the offices of the mining 
companies. The majority of the white miners 




are of English descent, largely coming from j. M. jones, Manager, 



Premier Mine, 



the hematite mines of Cumberland, and the tin, 
lead, and copper mines of Cornwall. They come to the fields 
in search of employment, which is given as occasion arises. 
Experience in other kinds of mines is soon adapted to the 
conditions in the Diamond Fields, and the men in the De 
Beers mines show a high average of efficiency. The nationalities 

407 



4 o8 THE DIAMOND MINES OF SOUTH AFRICA 

of the mechanics, engine-drivers, and others working about 
machinery are Scotch, English, and colonial, with a sprinkling 
of Americans and other nationalities. Those working on the 
floors and about the washing machines are largely of colonial 
birth English and Dutch, the balance being mostly home- 
born Englishmen. 

The majority of the white workers above and below ground 
have their homes in Kimberley and the other neighboring min- 




The Engineers, Mechanics, and Workmen who built De Beers Crushing Plant. 

ing towns. Wages paid to European day laborers on the surface 
range from IQJ. to 15^. a day; mechanics receive higher pay, 
which ranges from i6s. %d. to i per day, and white miners are 
paid the same rate. Miners who prove their competence are 
given contracts for specified work, by which their earnings 
are usually materially increased. Since 1892 all underground 
work has been done by the men working eight-hour shifts. 
The length of the working day above ground varies with the 
class of work done. Engine-drivers and men employed in gen- 
eral service at the mines work from ten to twelve hours daily. 



THE WORKERS IN THE MINES 



409 




On the depositing floors work begins in the summer at six 

o'clock in the morning ; time is given for breakfast, which is 

brought to the men, an hour's rest is allowed 

at noon, and work generally ends between 5 

and 5.30 in the afternoon. All mechanics work 

54 hours in the week, stopping at i o'clock on 

Saturday, at which hour all work on the surface 

ends for the day. Sunday is a full holiday 

above and below ground for every one except 

those in charge of pumping engines, pumps, Frank Manil X' Mana- 
ger, De Beers Mine 

boilers, man cages, etc., which must have atten- Compound. 
tion on Sundays as well as week days, and a few hands employed 
underground on necessary repair work to the shafts and mines, 
which cannot be done during the week while the mines are in 
full work. Kxtra time is allowed mechanics, miners, and others 
working under exceptional conditions. The pay of the men 
enables them to live comfortably in the mining towns, and as 
they are little given to dissipation, the thrifty are enabled to add 
to their savings yearly, as the work, except for the interruption 
by the war, is continuous and regular. 

Employes' houses in Kimberley are scattered through the 
city, and many of them own their own homes. Some of the 
miners' houses cost ^500 or over. They are commonly made 
of brick, or with corrugated iron sides and roofs the division 
walls being of unburnt brick and the outside walls being of the 
same material. The rental of a house in town 
ranges from 4. to _8 per month. The price 
of board at the boarding houses is about 25^. 
per week. The price of meat has commonly 
been about 6d. or ^d. per pound, although since 
the war, and owing to the devastation caused 
by rinderpest, the price of beef has nearly 

R. G. Scott, Superin- j 11 j T-> 11 . j r 

tendent, De Beers doubled. 1 o supply the urgent demand tor 

convict station. c h e aper meat, the De Beers Company has 

erected large cold-storage plants at Cape Town and Kimberley, 

and is now importing meat for sale to butchers at Kimberley. 







4 io THE DIAMOND MINES OF SOUTH AFRICA 

Beef and mutton make up the bulk of the meat sold. From 
March to August the markets are well supplied with game, 
chiefly springbok, stembok, guinea fowl, partridges, bustards, 
korhaan, and sand-grouse. Vegetables of all kinds are fairly 
plentiful and to be had at reasonable prices. For potatoes the 
current charge is from 155. to yos. per sack of somewhat less 
than 200 pounds. Cabbages, cauliflower, beets, beans, parsnips, 
carrots, onions, sweet corn, and celery are among the vegetables 
chiefly sold. Melons and fruits of all kinds are also plentiful in 
season. All vegetables and fruit brought from the neighboring 




Rinderpest. 

farms to Kimberley for sale are taken to the market square and 
sold under the supervision of the market master to green- 
grocers, East Indian hawkers, and the public generally. Flour 
has nearly a fixed value, being cheaper when the production 
in Basutoland and other grain-producing districts is plentiful, 
but never exceeds a certain price, fixed by the competition for im- 
ported flour upon which the government levies a duty. The flour 
chiefly used by the natives and by many of the white people as well 
is what is called Boer meal, which makes a brown bread, for only 
the bran has been removed. There are a number of roller mills 
in the country that produce flour which compares favorably with 



THE WORKERS IN THE MINES 



411 




412 THE DIAMOND MINES OF SOUTH AFRICA 

imported flour. There is an understanding between the Govern- 
ment, the local dealers, and De Beers, that De Beers Company 
shall only sell the necessaries of life to the natives in the com- 
pounds, and that the price shall range about the same as local 
prices in town. Any profits derived from these sales is to be 
distributed among public institutions and charities. 




Chiefs of the Batlapin Tribe. 

In the mines operated by the De Beers Company alone, 
more than eleven thousand African natives are employed 
below and above ground, coming from the Transvaal, Ba- 
sutoland, and Bechuanaland, from districts far north of the 
Limpopo and the Zambesi, and from the Cape Colony on the 
east and the south to meet the swarms flocking from Delagoa 
Bay and countries along the coast of the Indian Ocean, while a 
few cross the continent from Damaraland and Namaqualand, 
and the coast washed by the Atlantic. The larger number are 
roughly classed as Basutos, Shanganes, M'umbanes, and Zulus, 
but there are many Batlapins from Bechuanaland, Amafengu, 




THE WORKERS IN THE MINES 413 

and a sprinkling of nearly every other tribe in South Africa. 
Many travel hundreds of miles, and some more than a thousand 
miles, in order to reach the Diamond Fields, and many of these 
arrive half starved, and so weak and emaciated that they are 
almost worthless as laborers for weeks afterward. The natives, 
as a rule, are generally muscular, sinewy men, but not fleshy. 
Their feet are broad and flat, but their legs and arms are com- 
monly well rounded, and their thigh and shoulder muscles are 
large. The living skeletons who come in from the far interior 
districts of Africa gain flesh, as rapidly as lean cattle do in green 
pastures, when they reach a field flowing with meat and por- 
ridge. In the early years of the mines, the raw recruits were 
hooted at and sometimes pelted with stones by their kinsmen 
at the mines, as before noted, but of late years this rough greet- 
ing and hazing has very largely passed away. 

For the lodging and feeding of this great force of native 
Africans, special provision is made by the erection of large 
walled enclosures, called compounds, at the mines and on de- 
positing floors. There are seventeen of these compounds on 
the Diamond Fields, twelve of which are owned by the De 
Beers Company. The largest of all is the one at De Beers 
mine, and the description of this will serve for all, as they are 
essentially alike, except in size. 

Fully four acres are enclosed by the walls of De Beers Com- 
pound, giving ample space for the housing of its three thousand 
inmates, with an open central ground for exercise and sports. 
The fences are of corrugated iron, rising ten feet above the 
ground, and there is an open space of ten feet between the fence 
and the buildings. At the northern end of the compound there 
is an entrance gate. Iron cabins fringe the inner sides of the 
enclosure, divided into rooms 25 feet by 30 feet, which are 
lighted by electricity. In each room twenty to twenty-five 
natives are lodged. The beds supplied are ordinary wooden 
bunks, and the bed clothing is usually composed of blankets 
which the natives bring with them, or buy at the stores in the 
compound, where there is a supply of articles to meet the sim- 



4H THE DIAMOND MINES OF SOUTH AFRICA 




THE WORKERS IN THE MINES 




__^ U 



416 THE DIAMOND MINES OF SOUTH AFRICA 

pie needs of the natives. Besides these stores there is a hospital 
and dispensary, where any needed medical attention is promptly 
given, and a church for religious services, conducted by mission- 
aries delegated by the various church denominations. During 
week days this church is also used as a school for the instruc- 
tion of the natives. Compartments, with entrances opening 
through the walls, are set apart for latrines, and cared for with 
strict attention to sanitation. In the centre of the enclosure 
there is a large concrete swimming bath, in which most of the 
natives are at times found diving and swimming, as is vividly 
shown in the accompanying illustrations (see also page 440). 
If any fail to show the necessary regard to cleanliness, they are 
compelled to keep themselves clean. 

A competent manager is in charge of the compound, and 
his assistants are intrusted with the charge of preserving order 
and enforcing the compound regulations. The natives look 
upon the manager as their great white chief. He settles any 
disputes which may arise among them, and in conjunction 
with the mine manager investigates any complaints in reference 
to the amount of pay which has been allowed them, or any 
punishment or ill treatment by their white " baases," which, 
needless to say, is contrary to the regulations. 

The compound is lighted by electricity, arc lights being 
hung within and without the enclosure. When a newcomer or 
a number of natives, for they usually come in little troops, 
apply at the gate of the compound for employment, the appli- 
cants are admitted into the compound only by the immediate 
direction of the manager or his assistants. As soon as they 
enter, their clothes are searched to prevent the smuggling in of 
liquor, playing cards, or other forbidden articles; then the 
officer in charge of the dispensary examines each separately and 
carefully. No diseased man is given work, and any suffering 
from contagious diseases are sent at once to a quarantine build- 
ing outside the compound, where a temporary provision for 
such cases has been made. Within twenty-four hours, a second 
examination of every one admitted who shows any symptoms 



THE WORKERS IN THE MINES 



of disease is made by a physician in the employ of the company, 
who daily visits the compound. 

To enter the service of the company, each applicant must 
sign a written contract, binding himself to live in the compound 
and work continuously and faithfully for a period of at least 




De Beers Compound. 

three months, or longer if he so desires. At the expiration of 
a contract, the applicant may leave if he chooses, or his contract 
may be renewed indefinitely. Some of the natives in De Beers 
Compound have been employed continuously for ten years or 
more in the service of the company, for the more industrious 
prefer the certainty of wholesome food and steady pay to the 



418 THE DIAMOND MINES OF SOUTH AFRICA 




THE WORKERS IN THE MINES 



419 


















u 

H 




2 z 




r D 


1 


z 1 




8 




O W 




o9 




5 D 




w o 




o 




u 




H 




a 




o 




z 




o 




X 




t/1 




Q 




D 




2 




s 




o 




u 


j 


a 




Id 




a 




03 




H 




Q 



420 THE DIAMOND MINES OF SOUTH AFRICA 

shifting to any other occupation that is open to them, or to 
return to their old savage life. All contracts are filled out in 
behalf of the natives by an officer delegated for this purpose by 
the Registrar of Natives, a Government Official, in order to keep 
a record of all additions to the inmates of the compound, and 
provide assurance that the contract is signed with a full under- 
standing of its provisions. In consideration of this service the 
native pays a registration fee of a shilling, and a shilling per 
month during the term of his employment. All receipts from 




De Beers Compound Musical Band of Natives. 

this source, except the registration fee, go to the Kimberley 
Hospital Fund for the care of sick and wounded natives. As 
the company provides for the natives in its own hospitals, where 
free medical attendance and nurses, as well as free food, are 
furnished, the Kimberley hospital receives a very large monthly 
contribution without being at any expense for the care of sick 
natives in the compounds. After his signature or mark has been 
affixed to this agreement, a native cannot leave the compound 
until the specified term has expired, except by the permission of 
the compound manager, which is rarely given because of the 
opportunities that would be opened for taking out diamonds. 



THE WORKERS IN THE MINES 



421 




A Fireside Gathering, Kimbcrley Compound. 

Underground work in the mine is carried on both day 
and night by three shifts, under the supervision of the mine 
manager and overman and three assistant overmen, one of whom 
is detailed to take charge of each shift. The shaft is reached 
through an underground passage leading from the compound, 
and a partition in this passage gives separate entry and exit 
ways to and from the mine. All laborers are taken up and 
down the shafts in cages. Each " boy " wears a number on his 
wristband for easy identification, and when he passes into the 
mine his number is taken by a guard, and a tally machine 
records each native as he leaves the compound to go to work ; 
on his return, daily, he brings a ticket noting in what working 
gang he was employed and what pay he had earned for the day. 
The natives commonly work for the contractors, who mine and 
tram the diamond-bearing ground at a price per load which is 
arranged by tender, and the natives are paid a fixed wage per 
diem ; but a worker must drill a certain number of feet of holes 
for blasting, which in soft ground is about twelve feet, or he 
must load a fixed number of trucks, in order to earn his daily 
pay. The natives usually work in the mines in gangs num- 
bering from ten to thirty men and boys. The limit of age 



422 THE DIAMOND MINES OF SOUTH AFRICA 

for the employes in the mines is fixed by government regu- 
lation, which provides that no boy under twelve shall be 
employed. Another regulation prohibits the employment of 
females in mining work. It is further provided that no native 
shall be employed underground, or in any of the compounds, 
except under the responsible charge of a white employe of the 




Natives making Coffee, Kimberley Compound. 

company. The handling of the dynamite cartridges used in blast- 
ing is intrusted solely to white employes, and all work done by 
the native gangs is laid out and directed by white overseers. 

The drilling in the blue ground is done for the most part 
with long hand drills, jumpers, which are sharpened at both 
ends, and which the natives readily learn to use effectively; where 
the blue rock is hard, the natives use single hand hammers. 
Their sinewy frames and powers of endurance enable them to 



THE WORKERS IN THE MINES 



423 



labor day in and out without any apparent injury to their health. 
As a matter of fact, nearly all gain strength and flesh in the 
mines. All the "drill boys" in De Beers mines are now 
natives, and are scattered through the mines on various levels 




Open Mine Workers, Kimberley Compound. 

while working, the number at any one point depending upon 
the size of the working face or stope of blue ground. At points 
half a dozen boys may be working together with drills, indus- 
triously pecking away at the diamond-bearing ground. Natives 
are also employed in clearing away the excavated ground, and 
loading the trucks, which run on tramways to the hoisting shafts 



424 THE DIAMOND MINES OF SOUTH AFRICA 

when working on a main level, or to chutes on the intermediate 
levels. If the roofs of the levels were transparent and a view 
were possible of the workers, whites and blacks, toiling day 
and night in these underground passages and stopes, gleaming 
with the white rays of electric lamps, or plunged in darkness, 
only relieved by the flickering yellow flame points of straggling 
candles this vast underground hive of workers would be a 
greatly stirring and impressive sight. As it is, some conception 




Kimberley Mine Compound. 

of the great mine may be built up piecemeal in the mind's eye 
by combining the illustrations of the men at work which artists 
in the mines have been able to make, some of which are given 
in the pages of this work. 

There is a certain racial resemblance in the temperament, 
character, and often in the speech of all these native miners, but 
there are also marked tribal distinctions. The natives are clan- 
nish, and it is rare to see members of two different tribes lodg- 
ing together. " Boys " of the one tribe always prefer working 
together, and this natural liking is humored to some extent in 



THE WORKERS IN THE MINES 



425 




Natives drilling in the Open Mine. 



selecting gangs to work, although the mixing of the tribes in the 

mines is inevitable, and often desirable. The Zulu, sprung from 

the warlike tribes 

moulded by 

Chaka, is one of 

the best of the 

native workmen, 

tall, straight, and 

erect in bearing, 

proud of the tribal 

traditions of the 

Ama/ulu, " the 

people of the 

sky," and, but for 

an exceptional fit 

of passion, a good- 
tempered, cheery, 

and ever willing and capable worker. The Amashangaans, com- 
ing chiefly from Portuguese East Africa, are closely akin to the 

Zulus, and resem- 
ble them in form, 
temperament, and 
working efficien- 
cy. The Trans- 
vaal Basutos rank 
with the other two 
as workmen, for 
most are indus- 
trious and capa- 
ble, and form the 
most obedient 
class of native la- 
borers, and nearly 

The Last of Open Mining in Dutoitspan Mine. & JJ lj eC Q me skilled 

in drilling. The men of most of the native tribes range over 
5 feet 8 inches in height. Many are fully 6 feet tall, and several 




426 THE DIAMOND MINES OF SOUTH AFRICA 



of the old hands are from 
6 feet 4 inches to 6 feet 6 
inches in height. To this 
high range the Batlapins 
from Bechuanaland are the 
most notable exception, for 
they often are not much 
larger than the dwarfed 
Bushmen of the Kalahari 
desert. They are not fa- 
vorites at the mines with 
the other tribes, or with the 
whites, for they are often 
impudent and meanly self- 
ish, and difficult to in- 
struct in underground 
work. 

The ordinary dress of the natives in the compound is a wool- 
len shirt, trousers, and shoes. They rarely wear any under- 
clothing, and when at work in the mines, a pair of ragged 




Native drilling Underground. 




Drilling Underground. 



THE WORKERS IN THE MINES 



427 



trousers, a blanket, or old breech cloth will often be their only 
covering. Occasional visitors to the mine are startled by the 
native disregard for cover ; but the natives are commonly alert to 
pass the word " umfas " (woman) from one to another when a 
lady visitor is seen in the mines, and then the native workers on 
the level ahead scramble for cover or hiding. 







The Midday Meal. 

When any injuries happen to the men from accidents in the 
mines, the suffering natives show remarkable fortitude in bear- 
ing pain and enduring the necessary surgical operations. Their 
blood is warm and pure, and cuts in their flesh, or bruises, heal 
very rapidly. They suffer most from diseases of the lungs, 
especially phthisis and pneumonia, which are common maladies 
of the native tribes outside of the mines, as well as within the 
compounds. They can readily obtain fresh vegetables and 
fruit, but the common choice of food, such as mealie meal and 
meat, exposes them to attacks of scurvy. In spite of the care- 
ful and repeated medical examinations before men are admitted 
to the compound, cases of leprosy are occasionally found. 
In such cases provision is made at once for the isolation of the 
sufferers. The Government officials are notified, and the dis- 
eased men are transferred to Robin Island, where the Govern- 
ment has a permanent leper station. Outbreaks of other 
contagious or infectious diseases are met by the isolation of the 
patients in a special lazaretto outside of the town, which is under 
the supervision of the board of health. Natives suffering from 



428 THE DIAMOND MINES OF SOUTH AFRICA 

any disease that is not infectious are cared for in the hospital 
of the compound, which has several wards, one for cases of 
fever, one for convalescents, and one for surgical treatment. 
A qualified dispenser is in charge of the hospital and dispen- 
sary, and physicians engaged by the Company are in daily 
attendance. 

At the shops in the compound any articles of food and cloth- 
ing which the inmates commonly want are supplied. The staff 
of life is corn, or mealie meal in some form, sometimes baked 
in hoe cakes, but generally made into porridge. A consid- 
erable quantity of brown bread made from Boer meal is also 
eaten, with meat, vegetables, and fruit in season. Meat is com- 
monly cooked by boiling or by roasting over wood fires. The 
prices are never permitted to be in excess of the common market 
prices in Kimberley. If a " boy " does not want the trouble of 
cooking for himself, he can buy ready cooked food, which is sup- 
plied by the company or at any one of a number of coffee shops 
in the compound. One of the favorite resorts belongs to a 
Zulu, popularly known as " Roast Beef," who had the misfor- 
tune to lose his leg in an accident in the mines. He does his 
cooking over an open wood fire with the aid of a few kettles and 
pans ; and a bare wooden table, usually made from dynamite 
cases, serves for his dishes ; but he is a chef in his line, in the 
eyes of the compound, and is making more money than he 
earned before he was crippled. 

There are a number of native tailors on the ground, who can 
fit and make a suit to order, or repair one, with no little dexter- 
ity. Native mining suits are usually made of the English cloth 
known as moleskin, and the tailors, in accordance with South 
African custom, put large patches on the seat and around the 
foot of the trousers. Sewing machines are commonly used, 
which the natives buy in Kimberley through the compound 
manager. Some work in the mines during the week, but like 
to earn additional shillings by cloth cutting and sewing during 
their leisure hours, when their machines may be heard clicking 
from morning till night. 



THE WORKERS IN THE MINES 



429 




430 THE DIAMOND MINES OF SOUTH AFRICA 

There are native barbers and hair-dressers, also, of whom 
the chief is " Sandy," a Cape boy, who struts about on Sunday 
in a khaki jacket with the airs of a tonsorial artist on the crest 
of fashion, and is reputed to make more on his holiday with his 
clippers than he can earn in a week with the drill below ground. 
He has not as much range for his art as a French barber, for 
most of his patrons want their hair cut off close to the scalp ; but 




Native making Bangles. 

. 



he is justifiably vain of the speed with which he lops off one 
bushy head of hair, and makes room for the next to fall. 

Pedlers of all sorts, dealing in cakes, tobacco, and ginger beer, 
have their stalls in the moving throng, especially on Sundays 
and other holidays, and here and there are to be seen workers 
in Kafir adornments, principally in armlets or bangles, and bands 
for the legs. These are usually made of fine copper and brass 
wire rolled upon rings of horse hair. The rings are about one 
eighth of an inch in cross section and from four to five inches 



THE WORKERS IN THE MINES 



431 




432 THE DIAMOND MINES OF SOUTH AFRICA 

in diameter, varying with the size of the hands over which they 
must be slipped. The wire is wound round the hair very skil- 
fully. European visitors occasionally supply gold wire to these 
workers, which the natives wind around the hair centres into 
fanciful bangles, some of which are very pretty. 

All the workers in the compounds are supplied with Bibles, 
printed in various tribal languages, which the natives are taught 
to read by missionaries. At any and all times De Beers Com- 
pounds are open to these teachers, who are specially delegated 
by English and German missionary societies. 

When a " boy " is once moved to apply his mind to any 
study, he will commonly plod on persistently, and there is among 
the natives generally an unfeigned respect for teachers, and pride 
in the attainment of any advance in learning. There is only the 
crudest notion of religion in the minds of these negroes, and 
the missionary must have unwearied patience who seeks to 
impress them with the idea of an invisible, omnipotent, omni- 
present God and Father of all. It is very difficult for the mis- 
sionaries to prove by the Bible that these savages should have 
only one wife, and this has been a great stumbling-block in 
teaching them Christianity. The native argues that, if he has 
only one wife,she is continually wrangling with him, but if there 
are two or more, they occupy themselves by wrangling with one 
another. And again, he says, the more wives he has, the more 
crops he can raise. The women do all the work at the kraals, 
and the men idle their time away in peace and plenty. 

The preachers at the compound chapel or elsewhere in the 
compound often call together their flocks with stirring notes of 
drum and trumpet, and at gatherings of natives lime-lights and 
lantern slides are also effectively used in vivid and telling illus- 
trations. Sometimes an interpreter stands at the preacher's 
elbow, to make his meaning clear to native listeners, for the 
tribal dialects in the compound are like the confusion of tongues 
in Babel. The missionaries are somewhat vexed by the Kafir 
" doctors," who keep before the natives the vision of old super- 
stitions, as they squat on the ground in the compounds, sol- 



THE WORKERS IN THE MINES 



433 



emnly laying out their "bones" and muttering incantations. 
They are so tricky with their impostures that it is difficult to 
bring any of them patently into contempt. 

Almost all of the natives are fond of sport. They have 
plays of various kinds which may be seen every day in the 
compound, but the chief show is naturally on Sunday, the holi- 
day for all. Then a number of the tribes put on their native 
dresses, and there are vivid 
spectacles of native dances, 
chants, and games. The Zu- 
lus often arm themselves with 
clubs or wooden assagais, or 
any long canes which they can 
brandish and strike upon their 
ox-hide shields, while they 
circle about in a ring, mark- 
ing time with a stamp of the 
foot that makes the earth 
quake. It is the traditional 
report that no one is admitted 
to this war dance who has not 
killed a man ; but the chances 
are that, in recent days, un- 
questionable evidence of this 
qualification is not strictly 
required. Nevertheless, the 
pretence of bloodthirstiness 
is very exciting, as warriors 
spring forward, one after another, swinging their assagais or knob 
kerries, and advancing their shields, while they show a pantomime 
of attack upon an imaginary enemy almost as vivid and thrilling 
as actual battle. When this dance begins, a circle of native spec- 
tators gathers about, shouting and crying with the passion of the 
scene, till the noise at times is deafening. Other natives, less 
particular than the Zulus, dance about in rings and crescents, 
waving any kind of stick in their hands, from a miner's candle- 




'Mshangaan in War Attire. 



2 F 



434 THE DIAMOND MINES OF SOUTH AFRICA 




THE WORKERS IN THE MINES 



435 



stick to a twig or old hatchet. Among these figure the fantastic 
Machopis, dancing to the music of native imbilas, or Basutos 
blowing their little reed or bone whistles and swaying about with 
strange contortions, accompanied by monotonous tapping on a 
crude drum made by stretching a raw ox-hide over the end of a 
barrel. The 'Mshangaans chant while dancing, but the Basutos 
are not gifted with 
musical voices and 
have no evident 
ear for music, al- 
though they are 
so fond of their 
own harsh and 
discordant blow- 
ing that they will 
pipe away on their 
hollow bones and 
dance for hours at 
a time on Sunday 
to their own pip- 
ing. 

Among the 
other native tribes 
there are many 
boys with fine 
voices, sweet 
toned or robustly 
sonorous, ranging 
from the highest tenor or falsetto to the deepest bass ; and some 
are readily trained to part singing. In De Beers and other of 
the larger compounds there are native choral societies under the 
charge of white instructors. The most popular songs are the 
familiar American negro minstrel and concert hall melodies. 
These are freshly ludicrous to one who pictures the black singers 
"climbing the golden ladder" and "wearing the golden slipper" 
on their big flat feet. The climax is reached when the high 




'Mshangaans in War Paint. 



436 THE DIAMOND MINES OF SOUTH AFRICA 

voices sing, " What are you goin' to wear ? " and the reply comes 
from the deep bass voices, " I'se goin' to wear a standin' collar." 
Native African chants are rarely heard in the compound, except 
sometimes as an accompaniment of native dances. 

At all hours of the day, until the stir and buzz throughout the 
big compound are hushed in the sleep of its thousands of inmates, 
the rattling and humming and squeaking of imbilas and gubos, 
and various other crude instruments of native fashioning, are to 
be heard, more or less widespread. The "imbila" is the same as 
the maninba noted by Dr. Livingstone in his travels in Africa. 
In the native villages it is made by fixing strips of board across dry 
calabashes. By grading the size of these gourds, different notes 
are produced when the overlaid strips are struck by a drumstick 
with an elastic gum knob. In the compounds empty dynamite 
boxes with tin cans fastened underneath the strips of wood sup- 
ply the lack of calabashes, and the striking knob is imitated by 
twisting a piece of rag tightly round the end of a stick. The 
native "gubo," as the Zulus call it, is an instrument also common 
throughout South Africa. This is a bow of bamboo with a tightly 
stretched string. The player holds the end of the bow against 
his parted lips with one hand and strikes the tight string with a 
slip of split bamboo. A peculiar effect is obtained in playing on 
this bow in the compound by attaching a calabash to the back of 
the bow, and holding this improvised sounding-board against the 
breast. These are the favorite instruments, but there are others, 
like the bone whistles of the Basutos, which are much cruder, and 
grate far more harshly on the ear of listening white men. 

That the native African has an inborn fondness for music is 
signally shown by its persistent pursuit in the compounds, even 
through refuse boxes and bones. It may advance in time, with 
education, to high artistic appreciation and accomplishment. 
Even at its present barbaric stage the Kafir may be greatly 
moved by the art of a great singer, as was evident when Madame 
Albani came to the diamond mines, for she never saw an audi- 
ence so passionately enraptured as the black men massed about 
her within the walls of De Beers Compound. 



THE WORKERS IN THE MINES 



437 




438 THE DIAMOND MINES OF SOUTH AFRICA 




A Quiet Game of Cards. 

There would probably be a common resort to gambling as 
well as to music, if the practice were not sharply restricted by 
the compound regulations and oversight. It was necessary to 
prohibit the playing of cards, because native sharpers were fleec- 
ing the tyros too unmercifully. There is still, probably, some 
covert card playing, for many of the natives understand a few 
of the games familiar to white men. Faro was played with the 
top of an empty dynamite box as a table, upon which cards were 
tacked. The game was probably introduced by natives from the 
Portuguese possessions. The native African has only a few 
games of his own devising. The most popular of these in the 
compounds, and Africa at large, is "umtshuba" or "chuba," 
the Syrian "mancala," or, as the Nubians call it, "Mungala." 
The widespread knowledge of this game is noted by Schwein- 
furth as one of the links of evidence of " the essential unity that 
underlies all African nations " ; and it has been shown by the 
investigations of Mr. Stewart Culin for the Smithsonian Institu- 
tion of the United States, 1 and by other reports, that the same 

1 "Mancala, the National Game of Africa," by Stewart Culin, 1896. 



THE WORKERS IN THE MINES 



439 




Natives playing Chuba. 



game with essential variations is played throughout Africa and 
extends along southern Asia as far as the Philippine Islands. 
For this game a long strip of board is provided, edged with two 
parallel rows of holes scooped in the wood. When a board can- 
not be procured, the rows of 
holes are made in the ground. 
The number of holes in a row 
varies widely, the Nubian 
"Mungala" having sixteen, 
while the board common on 
the Diamond Fields has from 
thirteen to seventeen holes and 
four rows. Each player has 
about two dozen pebbles in hand, and the play is in shifting the 
pebbles from one hole to another. Stanley calls the game an 

African " back- 
gammon," and 
speaks of the 
board as a 
"backgam- 
mon" tray. The 
word " mun- 
gala" is of Ara- 
bic origin, de- 
rived from 
" nagal," " to 
carry from one 
place to an- 
other." There 
is no apparent 
interest in the 
game to the 

ordinary white man's eye, but native players in the compounds 
and African negroes generally will keep on moving the little 
stones for hours at a time with evident satisfaction, taking up 
their opponents' pebbles, as certain combinations occur, until 




Natives playing Mancala. 



440 THE DIAMOND MINES OF SOUTH AFRICA 

one or the other has won all. The spectators usually offer 
advice to the players. 

There is some running, jumping, and wrestling in halting 
imitation of English athletic sports; and on special holidays, like 
Christmas, they have obstacle races, sack races, walking the 
greased pole, which lies horizontally over the swimming bath, 
and other comical features for the general amusement of the 
native and white spectators. But the workers in the mines are 
rarely nimble enough to figure with any distinction in these 
sports, and the only English games that can be called popular in 
the compounds are the counterfeit of cricket and football. The 
native wickets are made of empty paraffin tins, and the fine points 
of the game are not in evidence ; but there is plenty of hard 
swiping and sharp bowling, to the delight of the native players 
and the spectators. Christmas is the great holiday of the year 
for all, for everybody in the compound then receives for his 
Christmas box a loaf of bread, a bottle of ginger beer, and a piece 
of meat, and sports of various kinds are specially provided for 




Swimming Balh, De Beers Compound. 



THE WORKERS IN THE MINES 



441 




Natives smoking Indian Hemp. 

their amusement. Grabbing and diving for money thrown into 
the swimming bath by the directors and managers form a lucra- 
tive sport for the natives, and amusement for the lookers-on. 

At every gathering for dances, sports, or games of any kind 
there are more lookers-on than participants, for the African 
dearly loves a spectacle of any kind, and is commonly well 
pleased to stand or loll on the ground where he can get a view 
of the contributors to his entertainment. Some of these indolent 
ones will be smoking cheap cigars, and more rarely pipes. A 
native usually puts the lighted end of a cigar in his mouth, inhal- 
ing and blowing out the smoke, and shifting the hold of his teeth 
as the tobacco burns. Sometimes sets of boys are seen squatting 
on the ground and passing from mouth to mouth a lighted pipe 
filled with dry dagga, a native herb similar to the Indian hemp, 
that burns with pungent and stupefying fumes. The natives 
inhale the smoke, and, after a few puffs, a fit of violent coughing 
comes on which brings tears to their eyes. The use of this herb 
is not so extended as to cause any serious ill effect, but the native 
becomes stupefied for a time, though he soon recovers. 



< 
442 THE DIAMOND MINES OF SOUTH AFRICA 

There is some tribal jealousy and vanity, but the inmates of 
the compounds live together on good terms, as a rule. In their 
occasional fights they use bottles or stones or clubs, or anything 
they can lay their hands on quickly ; but, as soon as the guards 
come up, they hurry off to their rooms, where they are put 
under strict oversight for a time. Even these short encounters 
often leave many with sore heads and bruised bodies. Only 
once has there been the threat of a serious insurrection in the 
compound. This was at Christmas time, when the compound 
manager was absent for a few days. After the usual Sunday holi- 
day several hundred natives, chiefly from Kaffraria, refused to go 
to work on Monday morning, as the following day was Christmas, 
demanding the grant of Monday also as a holiday. I went at 
5 A.M. to the compound and urged the leaders of the strike to 
take their followers into the mines. The Basutos were willing 
to support me, and offered to drive the reluctant Fingos, or 
Amafengu, underground. After some protracted but ineffec- 
tive appeals, I sent word to Mr. Rhodes, who happened to be 
at Kimberley, that the Fingos refused to go to work, and sug- 
gested that he might come over and try his persuasive power 
on them. So he did, but after an hour of fruitless parley- 
ing we determined to try the demonstration of force, for the 
Fingos not only refused to work themselves, but barred the 
other natives from entering the mine. As they numbered from 
five to six hundred, they were rather a formidable barrier at the 
underground entrance. 

We then decided to call in the assistance of the police and 
our own guards, Mr. Rhodes riding to the police station, while 
I rode to a station where a number of extra guards were posted. 
When we came back into the compound with a force of fifteen 
men armed with carbines, the Fingos instantly began to pelt us 
with bottles and stones, and anything else which would serve as 
a missile. At this outbreak I asked the officer in charge to fire 
a few blank shots at the crowd of rioters, and in less than a 
minute there was not a native to be seen in the open area of the 
compound, for all scurried off like frightened sheep to their 



THE WORKERS IN THE MINES 443 

rooms. We then went around the compound, picking out the 
ringleaders, thirty-three in all, ranged them in line, and sent 
them to jail. They were soon brought up before the magistrate 
and each was fined ^3, which they obtained by a little beg- 
ging from their brothers in the compound. Meanwhile, it was 
difficult for us to restrain every native left in the compound 
from going to work that day on the first shift. 

After the ringleaders came back to the compound, they wanted 
a meal, but they were forced to go underground and work eight 
hours before any food was provided. Then they were singled 
out and led around the compound, one by one, as an exhibition 
or warning to others, before they were finally discharged from 
the employ of the Company and sent away from the works. 
One of our interpreters had been taken along with the rioters 
by mistake. He was so vociferous that some one put him in 
with the other noisy boys. A few days later, when I wanted an 
interpreter, the unlucky one said, " All right, Baas, I don't mind 
interpreting for you, but I don't want to be run in for it." 

No corporal punishment of the natives by white employers 
is allowed. If a boy is unruly, he may be placed in a room by 
himself until he can be taken to jail, and charged with whatever 
offence he has committed. The most common offence is petty 
thieving. There can be no doubt that the covert purloining of 
diamonds would be a frequent practice, and cause heavy losses 
to the diamond mining companies, if it were not for the com- 
pound system, which makes it impossible for natives to take any 
diamonds out of the compounds with them. 

A fine wire netting is stretched over the top of the com- 
pound to prevent the sly tossing of precious crystals over the 
walls, to be picked up by confederates outside the mining 
areas. Precautions are also taken to prevent the smuggling 
away of diamonds from the compounds, and all communication 
by the natives with persons outside the walls is carefully 
restricted. Until the expiration of his contract, no native can 
go through the compound gate, except by special permission, 
or when he is taken under guard before a magistrate for some 



444 THE DIAMOND MINES OF SOUTH AFRICA 



offence. If convicted, when his term of imprisonment expires, 
or after he has paid his fine, he must return to the compound 
and complete his contract. Before leaving the compound his 
clothes and person are thoroughly searched to prevent the dis- 
appearance of diamonds with them. Gems were sometimes 
found secreted in clothing, or shoe heels, or canes, or cans with 
false bottoms, in fact, in anything that the natives were allowed 
to take out with them. Even this close inspection did not bar 

the practice of steal- 
ing, and there was an 
inexplicable trickle 
of fine diamonds 
from unlooked-for 
quarters, until it 
became known that 
natives on the point 
of leaving the com- 
pound were swal- 
lowing diamonds 
and conveying them 
away. 

In 1895 one na- 
tive had the nerve 
and capacity to swal- 
low a lot of dia- 
monds worth ^750, 
and did not appear 
to suffer by this strain upon his digestion. There has been only 
one authentic instance where a native has embedded diamonds in 
his flesh this was done by a native in De Beers Convict Station, 
who made an incision under the shin bone and concealed several 
small diamonds wrapped in a rag. This native had symptoms of 
tetanus, and the visiting physician (Dr. Otto) searched the man's 
body, and, finding an ugly-looking wound on his leg, cut it open, 
and to his great surprise found a rag full of diamonds. The na- 
tive soon recovered, a wiser, if poorer, man. The largest yield 




Diamonds which a Native had swallowed, and which were 
recovered by the Guards in the Compound, 




DIAMONDS SWALLOWED BY A NATIVE AT ONE TIME, AND ALL RECOVERED 
AFTER FOUR DAYS. TOTAL WEIGHT, 348 CARATS; VALUE, 1067:4:6. 



446 THE DIAMOND MINES OF SOUTH AFRICA 

of diamonds which a native had swallowed is represented by 
the illustration on page 445, each diamond being drawn the 





Diamond Thief. 



Diamond Thief. 



exact size of the original. There is no apparent fear of 
swallowing any stone which can be forced through the throat, 





Diamond Thief. 



Diamond Thief. 

and in one instance a diamond as big as a large chestnut and 
weighing 152 carats was hidden for over seven days by this 
means. 



THE WORKERS IN THE MINES 



447 




Diamond Thief. 



The swallowing of a rough diamond is evidently so easy, 
but so difficult to detect, that it was necessary to put an end to 
the practice by providing a longer 
period of detention and search. 
At the close of their contracts, na- 
tives whose terms of service have 
nearly expired are placed together 
in a commodious room capable of 
holding two hundred men or more. 
They enter this room entirely 
naked. Their clothes and bag- 
gage are deposited in sacks marked 
in accordance with the number on 
the arm band. Blankets are sup- 
plied for clothing, and as wraps 
when sleeping. They are fed, and 
generally well cared for, free of 
cost to themselves. While in the detention room they are 
under strict supervision of white guards, so that any diamonds 
they may have swallowed must be left 
behind before they leave. Natives have 
been known to keep diamonds in their 
bodies for over seven days. At the 
end of five days of detention, generally 
on Saturday morning, they are released. 
Meanwhile, the clothes placed in the 
sacks have been thoroughly searched ; 
and departing natives are not allowed 
to take away with them anything but 
soft goods. In fact, they are even re- 
quired to leave their boots behind, for 
cunning smugglers used to insert dia- 
monds in their boot heels so neatly that 
the trick could not be detected without 




Diamond Thief. 



cutting away the greater part of the sole of the boot. Boots and 
shoes, and other articles which are not allowed to be taken from 



448 THE DIAMOND MINES OF SOUTH AFRICA 




the compound, are sold or given away to customers or friends 
before their owners leave. 

It may be that De Beers Compound is a " Monastery of 
Labour," as was wittily said by a lady visiting the fields as a 

correspondent of the London Times ; 
but the testimony of all careful ob- 
servers on the ground affirms the 
beneficial effect of the restrictions 
from dissipation, and the general 
good cheer of the workers. Mr. 
Thomas H. Leggett, an entirely 
independent and competent Ameri- 
can witness, wrote of his inspection 
of the men in the compounds, 
in Cassier's Magazine, September, 
1898 : "These chaps are well cared 
for, contented, and happy, as proven 
by the fact that many have been 
there for years ; and the secret of it 
lies in their not being able to get drink." 

Occasionally a visitor at the fields is less observant and can- 
did. One such was a member of the Legislative Assembly of 
Cape Colony, who came to Kimberley to investigate the con- 
ditions of life and treatment of the natives in the compound. 
On arriving at De Beers Compound, in company with his wife, 
he first impressed upon the natives whom he met that he was a 
member of the Cape Colony Legislative Council. He had come 
to the fields in their behalf, and he wanted them to tell him 
freely everything of which they had to complain. With the aid 
of an interpreter he interviewed a number of natives in the com- 
pound, asking searching questions about their treatment. One 
native told him that he had been working for eight years in the 
mines and had been outside the compound only three or four 
times in all that period. When asked if he was well treated in 
the compound his answer was, " If I didn't like it, Baas, I 
wouldn't be here." The visitor's wife meanwhile kept tugging 



Diamond Thief. 



THE WORKERS IN THE MINES 



449 



at his coat continually, saying in Dutch, " They treat the Kafirs 
altogether too well here ; they will be spoiled by such good treat- 
ment as this." Before leaving, the legislator said that he was 
glad to have the opportunity to inspect fully the operations of 
the compound. From what he had heard he had been much 
opposed to compounds, but he now saw with his own eyes that 
he was wrongly informed, and henceforth he should be a strong 
advocate of the system. Yet a year or two later, when ques- 
tions affecting De Beers Company and the compound system 
arose in the Upper House, this gratified member was one of the 
first to denounce the system in an intemperate speech. 




Ue Beers Machine Shops. 



2G 



CHAPTER XV 



THE MINING TOWNS 




Kimberley 

IMBERLEY, the largest of the cluster of dia- 
mond towns on the Fields, is, like the rest, the 
natural efflorescence of the mines near which it 
is situated, and from which it derives its birth 
and being. Its mushroom growth must have 
withered like so many other pretentious upstarts 
from the mining fields, had it not been for the fact of its rising 
on ground of such sustained richness and promise. While the 
diamond-studded blue ground continues to show a persistent 
extension in depth and in richness, and while man's energy and 
art avail to pierce and extract it, the Kimberley of the surface 
will surely continue to flourish. 

It might indeed be said, without any stretch of imagery, that 
the modern Kimberley is literally as well as essentially built up 
on the yield of the mines. This has been brightly noted by the 
late Rev. James Thompson in his pleasing sketch of the modern 
Kimberley. "Kimberley, as we know it," he says, "with its 
streets and warehouses, and shops and schools and churches, is 
largely built upon that strange mixture known as debris, every 
atom of which has a story to tell if it could only speak. As in 
any English town you can go down foot after foot through the 
different strata representing the pavements or pathways upon 
which successive generations of ancestors pressed their feet ; so 
in Kimberley we have beneath the present surface of our road- 
ways the red soil on which our fathers pitched their tents, and 
which their labor soon covered up by spreading out all around 
them the heaps thrown out of that great hole which now looks 

45 



THE MINING TOWNS 



so desolate, but which was once the centre of activity and throb- 
bing life which made Kimberley famous throughout the world." ' 
Dr. Thompson marks the middle age of Kimberley as the 
period when decent buildings of iron and wood, with here and 
there more pretentious brick, had replaced the age of canvas ; 
but when there were no softening or beautifying surroundings, 
when every tree and bush had been cut down, and when the 
veld once dotted with thorn trees had become a vast expanse 




Snow in Kimberley, 1876. 

of wind-swept dust as gray as the iron dwelling places which 
alone seemed to convert the desert into a town. This was the 
period preceding the introduction of an abundant and pure 
water-supply that wrought such a transformation in the appear- 
ance of the city. Now the upspringing of flowers of varied hue, 
and green thickets and vines and trees in the gardens that now 
surround nearly every house in town outside the business 
quarter, has made during many months of the year a beautiful 
country town of the old and barren Kimberley. 

In spite of the visible yield of the mines and the consequent 
prosperity of the town there was, for many years, a prevailing 
1 Christmas number, D. F. Advertiser, 1898. 



452 THE DIAMOND MINES OF SOUTH AFRICA 




Theatre Roval. 



distrust of the permanency of the diamond-bearing deposits and 
the consequent stability and future of the city that was founded 
upon them. But later, as systematic development gave sub- 
stantial assurance of the endurance of the mines, the advance in 
the architectural beauty of the residences and public buildings in 
Kimberley has been marked. Now many of the residences of 
the more wealthy townspeople are not only substantial, but dis- 
tinctly ornate in character, with widespreading verandas rising 
in the midst of green lawns and lovely gardens. Some of the 
public buildings already erected or in process of erection need 
not fear comparison with any like structures in any city of its 
size in the world. 



THE MINING TOWNS 



453 



Among these structures is a handsome and well-appointed 
theatre, built of burnt brick with stone facings, excellently situ- 
ated for the accommodation of theatre goers. This building, the 
Theatre Royal, was designed to introduce all the latest improve- 
ments in theatrical construction, and its acoustic properties are 
particularly fine. The commodious stage has a face of 54 feet 
and a depth of 38 feet, and is so arranged that the whole stage is 




Theatre Royal Interior. 

in full view of the audience in the box stalls, dress circle, family 
circle, and gallery. The theatre is lighted by electricity, and its 
fire exits are so complete and well placed that in case of need 
the whole audience could leave in a very few minutes. 

The Town Hall is another building that deserves special 
mention. It was erected by resolution of the borough council 
on the Market Square after the destruction by fire of the old 
town hall. This building is designed in the Roman-Corinthian 
style and its appearance is notably pleasing. Its site is in the 



454 THE DIAMOND MINES OF SOUTH AFRICA 



centre of the Market Square, a particularly convenient position. 

There are three entrances to the main hall, which is finely pro- 
portioned, 105 feet in 
length, 50 feet in width, 
and 35 feet in height. At 
one end there is a stage 
25 feet wide, and a hand- 
some proscenium and 
space for the orchestra is 
also provided. 

There are emergency 

The Town Hall. , , 

exits opening into large 

yards that afford abundant protection in the event of the out- 
break of fire. Passages along the building lead to suitable ad- 
ministration offices for the borough engineer, market master, 





Dutoitspan Road, Kimberley. 



sanitary inspector, and native officials. At the back of the main 
hall extends the market house, over 83 feet wide and running 
the full width of the building. In the east wing of the building 
is a council chamber, 50 feet long by 26 feet wide, and, opening 



THE MINING TOWNS 



455 




Kimberloy Hospital. 

out of the chamber, rooms for the mayor and councillors. In 
the other wing of the building, accommodation is provided for 
the town clerk and his assistants. The building is substantially 
constructed of the best burnt brick covered with cement and 
enriched with cornices. 

On the site of the old Kimberley Hospital, established in 
1871, a new and spacious building has been erected, with sev- 




The Kimberley Club. 



456 THE DIAMOND MINES OF SOUTH AERICA 

eral outlying wards. The main building, about three hundred 
feet long, contains the operating rooms, convalescent room, 
and the Merriam, Victoria, and Lanyon wards for the reception 




Hall of Kimberley Club. 



of European patients only. The detached buildings comprise 
native medical and surgical wards, each containing fifty beds ; 
the Southey ward for colored women and children ; and isolation 



THE MINING TOWNS 



457 



wards for infectious cases ; male and female contagious disease 
wards, and mortuaries. The offices of the resident officials, a 
dispensary and doctors' quarters, nurses' home and chapel, with 
a further provision of European and native kitchens, make the 
hospital complete and comfortable. This hospital has accom- 
modations for 250 patients, European and colored, and from 
the day of its 
erection it has 
been of indispen- 
sable service. 
During the single 
year of 1897, 
2683 patients 
were admitted, 
798 of whom were 
Europeans, and 
the remainder na- 
tives and persons 
of color. Six hun- 
dred and sixty- 
three patients 
were admitted 
free, or on sub- 
scribers' letters. 
Besides this ser- 
vice it should be 
noted that the 
number of day 
patients treated during the same year was 1220; one of the hos- 
pital doctors is in attendance in the day-patients' room for an 
hour every morning to give advice without charge to the poor. 
To all who cannot afford to pay for treatment, medicines are fur- 
nished free. Every subscriber is entitled to give a letter of ad- 
mission to one patient for every 1 2s. subscribed, upon the sole 
stipulation that the person receiving the letter must be too poor 
to pay for his or her own treatment. The staff of the hospital 




Horns of South African Antelope. 



458 THE DIAMOND MINES OF SOUTH AFRICA 



consists of two resident house surgeons and a visiting body of 
seven local practitioners. The matron and forty-two nurses con- 
stitute the nursing staff. A recent addition has been made to 
the original hospital, in which will be the maternity ward, for the 
sake of providing the needed accommodation and the training 
of experienced midwives. The cost of this hospital with its 

enlargements has 
been upwards of 
/3 0,000. 

The Kimberley 
Club has a commo- 
dious and finely 
furnished house on 
Dutoitspan Road. 
This building was 
erected in 1896 on 
the ashes of two 
predecessors which 
had been unfortu- 
nately destroyed by 
fire. It possesses 
a unique collection 
of trophies of the 
chase, and its list of 
visitors bears the 
name of many of 
the most notable 
men in the British 
Empire. 

Besides these structures a government building of massive 
stone and brick on the north side of the Market Square de- 
serves mention as one of the conspicuous edifices in the city. 
Here the High Court of Griqualand is held. The magis- 
trates' courts are arranged on either side of the entrance, and 
rooms are provided for the Civil Commissioner, Judges, and 
Magistrates. 




Horns of South African Antelope. 



THE MINING TOWNS 



459 



The Kimberley Public Library is a well-built building, con- 
taining three large rooms, of which one is free to the public, and 
the others reserved for subscribers. Smaller rooms are provided 
for the librarian and committee. It is especially notable for its 
remarkable store of reference works, which is esteemed to be the 
best in South Africa. It contains in all twenty-two thousand 
books, many of 
which would be 
irreplaceable if 
destroyed. The 
building up of this 
library is justly 
credited to the 
fostering care of 
Mr. Justice Law- 
rence, the Judge 
President. 

Midway be- 
tween Kimberley 
and Beaconsfield 
stands the Kim- 
berley Sanatorium, 
a superb structure 
erected by the 
liberal contribu- 
tions of De Beers 
Consolidated 
Mines Limited at 
a cost, with its fur- 
nishings, of ^"26,000. Its fine enclosing grounds, the gift of 
the London and South African Exploration Company, were 
artistically laid out under the direction of Mr. Fenner of the 
De Beers Forestry Department. The larger part of the build- 
ing is designed for the accommodation of guests, and the smaller 
block contains the billiard room, smoking room, kitchen, ser- 
vants' and store rooms. The buildings are of burnt brick, two 




Horns of South African Antelope. 



460 THE DIAMOND MINES OF SOUTH AFRICA 



stories in height, with ample verandas and balconies ; all the 
rooms are large, lofty, and handsomely furnished, and in the 
construction the best sanitary knowledge has been applied. 
The building is lighted throughout by electricity, and abundantly 
supplied with pure water. 

The Masonic Temple was erected in 1889 by the combined 
lodges of the city. Its main hall is spacious and admirably 

lighted by elec- 
tricity, and the in- 
terior throughout 
is very handsomely 
decorated and fur- 
nished. At the 
top of the staircase 
there is one of the 
finest stained glass 
windows in South 
Africa, which was 
presented to the 
lodges by Mr. 
Rhodes. 

The post-of- 
fice, police bar- 
racks, and railway 
station have no 
special pretension 
to architectural 
beauty, but they 
are serviceable 
The offices 




Horns of South African Antelope. 

structures for the uses to which they are applied, 
of the De Beers Company are in the centre of the business sec- 
tion of the town, and are, as might be expected, excellently 
designed buildings, and stand out notably among the business 
edifices that surround them. 

The South African School of Mines was established at Kim- 
berley in 1896. The first two years' studies are taken at the 



THE MINING TOWNS 



461 



South African College, Cape Town, or at similar colleges at 
Grahamstown and Stellenbosch, the third year at Kimberley, 
and the fourth at Johannesburg. The object of the school is to 
train young men in South Africa as mining engineers. Suitable 
buildings were erected at Kimberley at a cost of ^9000, De 
Beers contributing on the pound for pound principle with the 
Educational Depart- 
ment of the Colony. 
There were twenty 
students in attendance 
during the year 1901. 
De Beers mines and 
workshops are open 
to the students, where 
they are given practi- 
cal instruction in min- 
ing and mechanical 
engineering. Their 
theoretical training is 
under the supervision 
of Professor J. G. 
Lawn, assisted by 
Professor Orr. The 
management of the 
school is entrusted to 
a local committee, 
consisting of the four 
members of Parlia- 
ment representing the 
Kimberley district, the member of the Legislative Council for 
Griqualand West, the Inspector of Mines, the Mayors of Kim- 
berley and Beaconsfield, the Chairman of the Public Schools 
Committee, and myself. I have the honor of being chairman 
of this committee. 

There are six distinct church establishments in Kimberley, 
the Anglican, Presbyterian, Baptist, Roman Catholic, Wes- 




Horns of South African Antelope. 



462 THE DIAMOND MINES OE SOUTH AFRICA 




Professor J. G. Lawn, Kimberley 
School of Mines. 



leyan Methodist, and Griqualand 
West Hebrew Congregation. The 
Anglican denomination has three 
churches in Kimberley, St. Cyprian's, 
St. Augustine's, and De Beers, besides 
churches at Beaconsfield and at St. 
Matthew's, Barkly Road. The larg- 
est church provides accommodation 
for 650 attendants. The first edifice 
of the Church of England was built 
at Dutoitspan, the pioneer town on 
the fields, and subsequently trans- 
ferred to Beaconsfield. When Kim- 
berley became the principal city of population, St. Cyprian's 
Church was erected on Church Street and removed to its present 
position in 1878. Kimberley became part of the diocese of Bloem- 
fontein, and 
gave its name 
to an Archdea- 
conry compris- 
ing Griqualand 
West and Be- 
chu analan d. 
The Archdea- 
con of Kim- 
berley is the 
head of the 
church organi- 
zation in this 
part of the dio- 
cese. 

The Kim- 

berley Presbv- Professor Orr, Kimberley School of Mines. 

terian Church was founded in September, 1877, and has over 
four hundred enrolled communicants and a still larger number 
of adherents. 




THE MINING TOWNS 



463 




Sir Alfred Milner passing the Offices of De Beers Consolidated Mines Limited on his First Visit 

to Kimberley, 




De Beers Offices decorated in Honor of the Governor's Visit. 



464 THE DIAMOND MINES OF SOUTH AFRICA 




Kimberley Public Library. 



In 1889 the Rev. James Hughes, of Port Elizabeth, at the 
invitation of the Baptist Union of South Africa, came to the 
Diamond Fields and held the first denominational meetings in 
the Good Templars Hall in Kimberley. Through his efforts a 




The Sanatorium. 



church was formed, and in 1892 the foundation stone of the 
present commodious Baptist Church in Dutoitspan Road was 
laid. 



THE MINING TOWNS 



465 



The foundation stone of St. Mary's Roman Catholic Church 
was laid on the feast of All Saints, 1879, by the Vicar Apostolic 
of Natal and Griqualand West. For many years previously 
a Catholic Church had been maintained on the fields, but its 
building was too small for the growing congregation. The 
foundations of the new building had just been completed, in 
August, 1879, when the old building was levelled to the ground 
by a terrific hail-storm. This was looked upon as a significant 




Masonic Temple. 

warning to replace the iron sides of the new church with brick, 
and the present edifice was accordingly erected, which will 
accommodate about five hundred people. It is adorned with 
stained glass windows, paintings, and oak altar and reredos, the 
gifts of its parishioners. 

Wesleyan Methodist missionaries were probably the first to 
visit the River Diggings in 1869, and representatives of this 
energetic denomination were among the first also that flocked 
to the Dry Diggings at Dutoitspan and Kimberley. The first 
regularly appointed minister of this church came in 1871, and 



2 H 



466 THE DIAMOND MINES OF SOUTH AFRICA 



the succession since that year has been unbroken. The Metho- 
dists erected their first church at the West End, but as the 
town moved eastward, a new church, Trinity, was built to meet 



. 









The Post-office, Kimberley. 

the call from that quarter. The original Trinity Church was 
blown down by one of the fierce gales sweeping over the Fields, 
but a second Trinity has now taken its place. There are now 
in Kimberley three Wesleyan churches for whites, two for 
natives, and one for other people of color, and a missionary 

is in daily attend- 
ance at the com- 
pounds. It is esti- 
mated that there 
are probably not 
less than three 
thousand persons 
under the charge 
of these seven min- 
isters. At a very 

Nazareth House, Kimberley. . , . . 

early date in the 

history of the Fields the foundation stone of the Hebrew 
Synagogue was laid on the Dutoitspan Road, occupying a site 
donated by the London and South Africa Exploration Com- 
pany. 




OF r . 

:./fO 4*2.- 



THE MINING TOWNS 



467 




The Author's House at Kimberley. Wistaria in Bloom. 




Another View of the Author's House and Garden. 



468 THE DIAMOND MINES OF SOUTH AFRICA 



Experience has shown that Kim- 
berley has special attractions as a 
health resort in spite of the occasional 
intense heat of its summer days and 
the blasts of its high winds laden with 
dust. It has the pure atmosphere of 
the high karoo plateau, and even in the 
hottest days the bright starlight even- 
ings are usually cool and refreshing, 
inviting the people to live for the 
greater part of the year in the open air 
on verandas and balconies. During 
the winter months the nights are often 
extremely cold, and well protected 
dwelling rooms are essential for com- 
fort and health ; but during the day the atmosphere is commonly 
clear, and so still that the severity of the cold is not felt, and all 
kinds of active outdoor exercise are agreeable in the bright sun- 
light of the unclouded skies. It is noted by the medical officer 
of health in Kimberley that the number of days of unbroken 




"Smell my Flowers." (The au- 
thor's daughter, Dorothy, and 
Jim, a good specimen of Basuto 
boy.) 




Kimberley Race Course. 



THE MINING TOWNS 



469 




miij!!!!}!!!!" 1 !'" 



The Grand Stand, Kiniberley Race Course. 

sunshine are particularly enjoyable to newcomers. They will find 
that the air they breathe is never heavy, damp, or oppressive, 
but always dry and light, and, outside of the centre of the town, 
pure and invigorating. The heavy thunderstorms that occa- 
sionally occur bring deluges of rain, but the water rapidly flows 
off the surface, and as vegetation is scanty, the soil remains 
exceptionally dry. 

It is this marked climatic attraction which, in connection with 
the pleasure resorts of the city, suggested the establishment of the 
Kimberley Sanatorium. 

Beaconsfield 

Dutoitspan, as before noted, was the original town on the 
Diamond Fields. When crowds flocked to the Fields and a 
demand for greater accommodation arose, the London and 
South Africa Exploration Company laid out the town of 
Beaconsfield, which adjoins Dutoitspan on the north. It was 
laid out as a business town, and has grown to be a place of 



470 THE DIAMOND MINES OF SOUTH AFRICA 

considerable size containing several thousand inhabitants. The 
town limits extend to the farm Dorstfontein, but the business 
and residence quarters are all within the farm Bultfontein. The 
main street in Beaconsfield leads direct to Kimberley. Many 
of the houses are of brick and iron, but the larger number are 
of unburned adobe brick, made of clay dug directly from the 
soil on which the house stands. With few exceptions all are 
unpretentious, one-story buildings. 

The town originally belonged to the London and South 
Africa Exploration Company, the organization which laid out the 
town, but together with all that company's property passed into 
the hands of the De Beers Company in 1898. According to 
the common practice houses are put up by the tenants on lots 
leased from the Company. Beaconsfield is laid out in wards, 
and has a distinctive Municipal Government of its own, consist- 
ing of a Mayor and Town Council and the usual town officers. 
The Mayor is a member of the Council and elected annually. 
Although Beaconsfield has thus a distinctive individuality, the 
business firms are very largely branches of corresponding firms 
in Kimberley. The town transacts considerable business, chiefly 
in stocks which are carried for the use of the mines ; but there is 
also a large number of shops which carry supplies of all kinds 
for the consumption of the white residents as well as for the 
native population which lives in locations near the town. 

Wesselton 

Close adjoining to Beaconsfield lies the little village of Wes- 
selton. This was laid out by the owner of the Wessels estate 
on Benaauwdheidsfontein farm. Its buildings resemble those 
of Beaconsfield, but are commonly of a poorer order of adobe 
brick structures, built like the Beaconsfield houses on leased lots. 
Wesselton has now only a few hundred inhabitants, mostly 
natives and East Indians. The natives are chiefly workers for 
debris washers about Dutoitspan and Bultfontein mines, while the 
East Indians are commonly kitchen gardeners and small shop- 



THE MINING TOWNS 



47 1 



keepers and pedlers. The various vegetables that are raised 
are sold in the little greengrocer stores, or hawked about by the 
pedlers in handcarts. Some of the East Indians also peddle 
clothing and knickknacks more or less industriously. 



Kenilworth 

On Kenilworth farm, about two and a half miles from Kim- 
berley, the so-called model village of Kenilworth is built. This 




Bachelors' Quarters, Kenilworth Village. 

village was planned in the latter part of 1888 by Mr. Rhodes, 
and laid out under his general direction by the late Mr. Sydney 
Stent, an architect then residing in Kimberley. It covers a 
space about half a mile long and a quarter of a mile wide, upon 
land owned by De Beers Consolidated Mines. 

The land was divided up into lots of about 80 by 100 feet, 
and upon these lots semi-detached houses were built, of brick 
with corrugated iron roofs, by De Beers Company. Nearly all 
of the houses are built of red burnt brick made at the brick- 



472 THE DIAMOND MINES OF SOUTH AFRICA 

fields in the neighborhood of Kimberley. The cottages rent 
from 1 IQS. to ^5 a month. The houses occupied by the 




Passenger Train for Workmen and their Families. 

unmarried men contain six rooms, and the other houses occupied 
by families contain from four to seven rooms. In the unmarried 
men's quarters each occupant has a room to himself. Nearly 




The Road to Kenilworth, and Kenilworth Reservoir. 

all of the houses are built with verandas, and all the lots are 
planted with fruit trees, vines, and flowers, supplied by the 
Company. Most of the residents take a keen interest in their 



THE MINING TOWNS 



473 



gardens and have added largely to their beauty by purchasing 
plants on their own account. In laying out the town, the vil- 
lage was originally planned with four main avenues, bounded on 
the north by a bordering avenue, on the south by the main 
road to Kimberley, and intersected by a central avenue. Only 
two of the avenues are at present completed. They are broad, 
well-made roads lined with blue and red eucalyptus, beefwood 
and pepper trees, and provided with wide sidewalks fronting 
the semi-detached villa-like residences. These avenues are finely 
macadamized and the streets watered by distributing carts. 




Kenilworth Club-house. 

Supply pipes are laid out along the streets and every garden is 
supplied with free water for irrigation from the Premier mine or 
Kenilworth reservoir. Separate pipes are laid to carry water for 
drinking purposes, and for this water a light charge is made, 
averaging about los. for 1500 gallons. 

A circle at the junction of No. i and Central Avenues 
divides the residences of the married people from the quarters of 
the single men, who occupy a row of houses on the south side of 
the circle in the heart of the village. One of the main houses 
on this circle is occupied by the Cape Government for a post- 
office, telegraph-office, and post-office savings bank. On the 
other side of the circle bordering on the central avenue is a club- 



474 TH E DIAMOND MINES OF SOUTH AFRICA 

house, a large brick building containing a reading room, dining 
room, kitchen, and manager's rooms. This building, like the 
residences, has a veranda in front, and is surrounded by trees. 
It is open to any white employe of the Company, but it is, of 




Kenilworth Village, the School in the Foreground. 

course, principally used by those living in Kenilworth. Citizens 
of Kimberley may visit it, and join in the social gatherings 
arranged by the residents in the village. On the north side of 
Central Avenue, opposite the club-house, is a schoolhouse con- 
taining three rooms, in which the library of the town is placed, 
and this is open after school hours for the distribution of books. 
The library has its own store of good books, but in addition to 
this stock, the Kimberley library contributes books by special 
arrangement, and it is practically operated as a branch of the 
Kimberley library. The school of Kenilworth is a primary 
school connected with the Kimberley public schools, and the 
children of the village are taught the usual elementary studies 
ranging up to the common English grammar school. When 
this grade is attained, arrangement is made for the attendance of 



THE MINING TOWNS 



475 



the children at the higher schools in Kimberley. To assist them, 
the Company provides free monthly tickets to and from Kenil- 
worth via the Kimberley-Kenilworth tram line. 

The village is wholly given up to residences ; there are no 
stores or shops of any kind. All supplies come from Kimberley, 
and by special arrangement the schoolhouse is used on Sundays 
for worship and mission work, and on evenings during the 
week by the various philanthropic and social organizations. 
The village is lighted by a few large arc lamps, and the houses 
by paraffin candles and kerosene. 

Arrangements are made by which the unmarried men take 
their meals at the club-house at a cost averaging about 255. a 




Kenilworth Village, with Meteorological Station on the Left. 

week. The men come in from their work to dinner, which they 
take in the dining rooms of the club ; their breakfast and lunches 
are sent out to the depositing floors or other places of work. 
The breakfast at the club is like that served at the better class 
of miners' boarding houses at Kimberley, consisting of bacon 
and eggs, chops, or steaks, or other substantial dish, bread and 



476 THE DIAMOND MINES OF SOUTH AFRICA 

butter, coffee or tea. Lunch consists usually of a meat dish 
with bread, vegetables, fruit, and tea or coffee. Dinner is the 
main meal, at which roast beef, roast mutton, and vegetables 
of all kinds are served. 

The shade and fruit trees of Kenilworth and adjoining plan- 
tations are the special pride of the village and of the De Beers 
Company, which has been indefatigable in introducing, acclima- 
tizing, and maintaining every variety that will thrive. Just 




View of Kenilworth Village. 

adjoining Kenilworth on the north is the orchard of the Com- 
pany, containing about 8000 trees, oranges, lemons, apricots, 
peaches, plums, pears, apples, quinces, and other fruits, as well as 
shade trees and grapevines. Most of the grapevines are trained on 
trellises. The first one built by the Company was 975, and the 
second 1 800, feet long. On these trellises all the best varieties 
of grapes are grown. The ripening season is from the end of 
December until the end of February, or during the summer 
months of a season stretching from October to May. Grapes 
and fruit from these orchards are largely distributed to employes, 




THE MINING TOWNS 



477 



and sent to hospitals and charitable institutions. Some fruit is 
sold in the compounds to natives at a price hardly reaching the 
cost of production. At times apricots have been sold at a shil- 
ling a hundred from the trees, and for sixpence when they were 
picked off the ground. In favorable seasons trees and vines are 
very prolific. 

The difficulties met with in raising fruit are frost in the early 
part of the season, when the trees are blossoming, and hail-storms 




Preparing Trenches for planting Vines and Trees. 

in the beginning of the year, when the fruit is young. Locusts 
come in millions and at times devastate the whole orchard, leav- 
ing the fruit exposed to the sun and at times badly eaten. There 
are two kinds of these locusts : one comes and stays for a day or 
so, doing what damage it can for the time being ; the other one 
alights on the trees for permanent occupation. They first ap- 
pear in the early spring as small insects. The little dark-brown, 
wingless creatures are commonly known as voetgangers (walkers), 
and come out of the ground when they are hatched, hopping 
along in countless myriads. The locusts plant their eggs in the 



478 THE DIAMOND MINES OF SOUTH AFRICA 

sands to hatch during the months of September and October. 
Sometimes all Kenilworth and the adjoining fields are swarming 
with these insects. In order to protect some of the gardens from 
young locusts, sheets of corrugated iron twenty-six inches wide 
are placed along, and leaning against, the fences. The locusts 
cannot climb up the smooth surface of the iron. In that way 
many residences are also protected. Sometimes servants are 
employed continually from morning till night in driving away 
the insects. They destroy all the vegetation over which they 
pass. The natives are very fond of eating them. They go out 
into the veld in large parties, and drive the voetgangers from 
all directions upon blankets, and then empty them into sacks 
which they carry to their huts. Flying locusts develop in about 
six weeks from the dark-brown little insects. The other variety 
that scourges the fields is a species of locusts with red wings, and 
their damage is the greater from the fact that they stay in one 
place until every green plant upon which they alight is destroyed. 
Swarms of these locusts occasionally appear, at times darkening 
the horizon, and following the wind. For the past seven years 
these swarms have been very troublesome. During one season, 
after consuming all the leaves, the leaf and fruit buds on the 
trees were entirely eaten off by these pests, destroying the fruit 
not only for that year, but for the following season. In spite 
of these drawbacks to fruit raising, the efforts of the Company 
have been unflagging. 



CHAPTER XVI 




FORMATION OF THE DIAMOND 

The Diamond-bearing Deposits 

VER the basin now extending as an arid karoo 
for hundreds of miles to the south of the Kim- 
berley Diamond Fields the waters of a great lake 
once spread. It is apparent that the diamond 
mines are on the northerly rim of this basin, 
for the beds of shale that everywhere under- 
lie the basaltic trap surface or country rock are notably thinner 
in the northern mine openings than they are farther south at 
Bultfontein and Dutoitspan, 1 and shortly after passing Kim- 
berley fields the shale terminates at the edge of the " bed rock " 
of the Vaal River diggings, an amygdaloidal trap which Dr. 
Stelzner' 2 determined to be olivine diabase. 

By the great open excavations and the extension of the un- 
derground workings, the rock formations of the karoo basin are 
very clearly revealed. The red soil that covers the surface of the 
country to the depth of from one to five feet is evidently the 
result of the decomposition of the friable face of the under- 
lying basalt, which is scattered in fragments over the country in 
jutting bouiders and rounded stones. This rock at De Beers and 
Kimberley mines is from twenty to ninety feet in thickness, but 
very much decomposed throughout. Below the layer is a bed 
of black shale, ranging in thickness from two hundred to three 
hundred feet. In this bed there is a considerable amount of 
carbon and a large quantity of iron pyrites. 

1 "Diamonds and Gold in South Africa," p. 19, Theodore Reunert, M.E. 

2 Dr. A. W. Stelzner, Professor of Geology at the Freiberg Mining Academy. 

479 



480 THE DIAMOND MINES OF SOUTH AFRICA 



SCALE 
100 



JOO 



KIMBERLEY 
FT. IN. 



Geological sections of Underlying the shale 

DE BEERS AND KIMBERLEY ROCK SHAFTS. . , . , . c . 

is a thin bed of conglom- 
erate, composed of small 
stones,some well rounded 
and others angular, and 
firmly cemented together. 
Its thickness, measured 
in the rock shaft in the 
Kimberley mine, did not 
exceed ten feet. This 
band has been styled by 
Professor A. H. Green 
the basement conglom- 
erate of the Kimberley 
shales, 1 and it is assumed 
by Mr. E. J. Dunn to 
be of the same origin as 
the Dwyka conglomerate 
belt on the northern base 
of the Zwarte Berg and 
Witte Berg mountains, 
forming the southern 
boundary of the old lake 
basin. 2 He holds that 
this conglomerate is a 
glacial deposit marking 
the shore line of the an- 
cient lake. 

Below the conglom- 

1 "A Contribution to the 
Geology and Physical Geogra- 
phy of the Cape Colony." 
Quar. Jour. Geol. Soc., Vol. 
44 (1888), p. 245. 
2 " On the Mode of Occurrence of Diamonds in South Africa." Quar. Jour. 
Geol. Soc., Vol. 30 (1874), pp. 54-59. 




FORMATION OF THE DIAMOND 481 

erate is a very hard amygdaloidal rock, called melaphyre by 
M. A. Moulle, 1 which was finally determined by Dr. Stelzner 2 
to be olivine diabase. Its mineral composition is the same as 
melaphyre, plagioclase, augite, and olivine, but one is granu- 
lar and the other porphyritic. It is about four hundred feet 
in thickness and is very hard. Underlying the amygdaloidal 
rock is quartzite, the thickness of which is not yet determined. 
The Kimberley rock shaft has passed through fourteen hun- 
dred feet of it, and the bottom of the shaft is still in the same 
formation. All these strata lie nearly horizontal, but dip 
slightly to the north. They are graphically presented in the 
sectional views of the rock shafts of De Beers and Kimberley 
mines shown on page 480. 

Through these layers of rock extend from an unknown 
depth the huge pipes containing the diamond-bearing deposits, 
or blue ground, which is a breccia filled with fragments of shale 
and other minerals. These immense funnels are obviously 
extinct craters filled with volcanic mud from below. All evi- 
dence to hand points to an aqueous formation, and the upheaval 
is shown by the upturning of the enclosing shales at various 
places in contact with the blue ground. 3 Many boulders are 
found in the blue ground of the same composition as the sur- 
rounding rock, but others have undoubtedly come up from 
greater depths than have yet been reached by the sinking of 
shafts. It is, however, highly remarkable that there was almost 
no apparent overflow in the filling of these craters, for the 
diamond-bearing ground is either level with the surrounding 
surface, or rises, usually, only a few feet above it in kopjes or 
hillocks. Outside of the mouths of the craters no diamonds 
have been found except at Dutoitspan, where the upheaval 
formed quite a hill, and some diamonds have been taken from 
the surrounding ground within a few yards from the margin of 

1 " Memoire sur la geologic generale et sur les mines de diamants de 1'Afrique 
du Sud." Annales des Mines, 8th Series, Vol. VII (1885), p. 193. 

2 Dr. A. W. Stelzner, Professor of Geology at the Freiberg Mining Academy. 

3 Still to be seen at De Beers Mine. 



z i 



482 THE DIAMOND MINES OF SOUTH AFRICA 




De Beers Diamond. Found March 28, 18 
Weight, 428} carats. 



the mine. It is also evident that the mines were not all fi led 

with the same material at one and the same time. Each mine 

has its distinctive character- 
istics, and even in the same 
mine all the blue ground does 
not seem to have been forced 
up at one time. This is par- 
ticularly demonstrated by the 
striking fact that, in both De 
Beers and Kimberley mines, 
the west side blue ground is 
wholly unlike the other por- 
tions of the mines, and carries 
fewer diamonds, and these are 
unlike the diamonds that are 
found in other parts. The 

blue ground which filled the west ends of these mines must 

have come up first, filling 

the whole crater. Afterward 

there was a second upheaval 

which filled the eastern parts 

of the craters with a richer 

deposit. The reason why 

the west end was not mixed 

with the better blue ground 

was because the west end 

parts of the mines formed 

benches, and were not ver- 
tically above the second 

boiling mass. Mr. Rhodes 

suggested this solution, and 

I quite agree with him. This Another view of the De 

peculiarity is noticeable in 

the other mines. 

The composition of the blue ground, which is the principal 

filling of the volcanic pipes, has been carefully determined by 




Diamond. After 
cutting, its weight was 228$ carats, and it was 
sold for ,13,600. 



FORMATION OF THE DIAMOND 



483 



Dr. Stelzner. This ground, he says, must be designated as a 
breccia. Most of the small or large angular-edged or rounded 
fragments of this breccia are composed of a green-black or blue- 
black serpentine-like mass. Fragments of rock which are found 
in the karoo formation, such as sandstone, shale, and diabase, 
are to be found in the blue ground. There are also other rocks 
in the shape of boulders of greater or less size, which are not 
known in the karoo formation, and have doubtless come from 
a much greater depth than the karoo beds, possibly from rocks 
upon which these beds lie. The mass of blue ground consists of 
olivine more or less changed 
by oxidation, with the follow- 
ing minerals : chromic diallage, 
bronzite, pyrope containing 
chromium, flesh-colored zir- 
cons (locally called Dutch 
boart),cyanite,biotite,chrome, 
titanium, and magnetic iron, 
and also small crystals of 
perofskite. 

In the Jagersfontein blue 
ground corundum is said to 
have been found. This was 
for a time held to be cordierite. 
The existence of small crys- 
tals of tourmaline and rutile is also reported. Professor J. G. 
Lawn, Kimberley School of Mines, reports that he discovered 
rubies and sapphires of inferior quality in the Frank Smith mine 
near Kimberley. Iron pyrites and barytes are found in the 
deposit resulting from washing the blue ground. The pyrites 
come from the country rocks, and become mixed with the 
diamond-bearing ground during the process of mining. The 
barytes is a secondary formation of small veins in the blue 
ground, or at its junction with the country rock. Beautiful 
crystals of doubly refracting or Iceland spar are occasionally 
found also near the junction of the blue ground and the rock. 




The Largest Diamond ever found in the Kim- 
berley mine. It weighed 503 carats, but was 
full of spots. 



484 THE DIAMOND MINES OF SOUTH AFRICA 

In Professor Lewis's discussion of the genesis of the diamond 
in 1886, he designated the blue ground variously as "dunite 
porphyry," "Saxonite porphyry," and " diamantiferous perido- 
tite." His application of the term " Kimberlite," now generally 
accepted by geologists, first appears in his paper of the following 
year, 1887, at the British Association meeting at Manchester. 1 
Dr. Stelzner thought this name should be adopted as concisely 
covering "a porphyritic volcanic peridotite of basaltic structure." 

In the mass of diamond-bearing blue ground in De Beers 
mine there is a curious dyke of igneous rock which extends from 
the southeast part of the mine around the east and north sides, 
and is lost in the unexplored poor blue ground of the west. 
Owing to its taking a serpentine course across the mine, it has 
received the local name of " snake." The upper end of this 
snake is at or near the surface, and the body extends down to 
the lowest workings. It does not adhere to the blue ground, 
and is very easily separated from it. It stands like a vein, 
nearly vertical, varying in thickness from two to seven feet. No 
diamonds have been found in it, yet Dr. Stelzner's investigations 
show that its composition is substantially the same as the sur- 
rounding breccia. It was difficult to obtain slides of the blue 
ground for microscopical observations and comparison, but 
after many trials Dr. Stelzner succeeded in getting a few sections 
which revealed these interesting facts : 

" The main body of the blue ground is entirely analogous to 
the snake rock, naturally more decomposed, but in essential 
points the microscopic features of blue ground and snake (not 
taking into consideration the numerous little slate fragments in 
the blue) are in an extraordinary degree alike. It therefore 
impresses upon one's mind that the "snake" is a younger erup- 
tive formation coming from the same volcanic source as the blue 
ground." 2 

1 "The Matrix of the Diamond," Henry Carvill Lewis, M.A., F.G.S., 
Professor of Mineralogy in the Academy of Natural Sciences, Philadelphia, U.S.A., 
at meeting of British Association at Manchester, August and September, 1887. 

2 Letter of Dr. Stelzner addressed to Gardner F. Williams. 



FORMATION OF THE DIAMOND 485 

On the looo-foot level of Kimberley mine a tunnel driven 
in the quartzite outside the margin of the mine shows several 
dykes of similar rock. Wherever these dykes exist there is a 
considerable quantity of water at the junction of the dykes and 
quartzite. 

There was a large mass of country rock in De Beers mine, 
which in the upper levels covered several claims, or approximately 
an area of 3000 square feet. It continued down to a depth of 
about 750 feet. It was an olivine diabase, and was the same as 
the amygdaloidal rock, except that it was filled with numerous 
veins of zeolites. The " Island," as it is called, was a gigantic 
horse of country rock embedded in blue ground, and has disap- 
peared in depth. Islands of the same rock appeared in the Kim- 
berley mine near the surface and at a depth of 1200 feet, and near 
the surface in Dutoitspan, Bultfontein, and Premier mines, where 
they have been left standing as the blue ground which sur- 
rounded them has been removed, and form huge islands in a 
sea of blue ground, which are locally known as Mount Ararats. 

Floating shale appeared at or near the surface of the mines 
and covered many claims. This was originally volcanic mud, 
and it contained no diamonds. It gradually became smaller in 
depth, and has disappeared in the lower levels. 

In the early descriptions of the mines fossil wood and plants 
are reported to have been found in the blue ground. I am of 
the opinion that these came either from the shale surrounding 
the craters, which was constantly falling into the open mines, or 
from the pieces of shale which became embedded in the blue 
ground at the time the craters were filled. The only fossils which 
have been found in the mines since they have been under my 
management are the fish which are shown in the illustration on 
page 486. They are embedded in sandstone which was found 
on the 1 85-foot level of Premier mine. 

The surface shales and basalt surrounding the pipes are 
called reef, and the masses of shale and igneous rocks, scattered 
through the blue ground in the upper levels of the mines, are 
commonly spoken of as floating reef. 



4 86 THE DIAMOND MINES OF SOUTH AFRICA 

After careful microscopical observations, Dr. Stelzner and 
others have reached the conclusion that the blue ground is of 
volcanic origin, and was forced up from below. This conclusion 
accords with the opinion which I formed of the origin of the 




Fossil Fish from Premier Mine. 



diamond-bearing deposit, during my visits to the Diamond Fields 
in 1884 and 1885. I then thought that the filling of the pipes 
was due to aqueous rather than igneous agencies, possibly to 
something in the nature of mud volcanoes. 



The Genesis of the Diamond 

The chemical composition of the diamond has long been 
determined, at least approximately. Sir Isaac Newton conjec- 
tured it to be of vegetable origin and combustible, but it was not 
until 1694 that Newton's assumption of its combustibility was 
actually proved by the famous burning glass experiment of the 
academicians of Cimento, at the prompting of the Grand Duke 
Cosmo III. 

Lavoisier, Guyton de Morveau, and others practically deter- 
mined, later, that the burning of a diamond with a free supply of 
oxygen converted it into carbon dioxide ; and, finally, the ex- 
periments of Sir Humphry Davy, in 1816, showed that the 
diamond was almost entirely pure carbon. Davy's conclusions 



FORMATION OF THE DIAMOND 487 

have been confirmed by Dumas, Stas, Friedel, Roscoe, and 
other eminent chemists, apparently fixing with extreme precision 
the chemical composition of the diamond. It is, however, note- 
worthy that the diamond is a non-conductor of electricity, while 
graphite and charcoal, substances so closely similar in chemical 
composition, are excellent electrical conductors. By the applica- 
tion of friction the diamond can be positively electrified, but 
Streeter says that it loses its electricity completely in the course 
of half an hour. 1 

So much, it may be claimed, we know ; but the process of 
the formation or crystallization of the diamond carbon is still 
uncertain. The proofs are most conclusive that the diamonds 
in the South African mines were not formed in situ, but have 
come up from below with blue ground. The frequent occur- 
rence of broken crystals embedded in the blue is sufficient 
evidence that the diamonds are not in their original place of 
crystallization, for it is impossible for nature to produce a frag- 
ment of a diamond. 

The late Dr. W. Guybon Atherstone, F.G.S., whose identifi- 
cation made known the first diamond of the South African 
Fields, presented his theory at a meeting of the South African 
Geological Society, as follows : 

" The succession of the strata in the Kimberley mine is pre- 
cisely the same as that of the lacustrine sedimentary beds, begin- 
ning from the quartzite base of the carboniferous rocks and 
shales, through the ecca and karoo formation, the coal-bearing 
shales of the Stormberg, to the dolerite, capping and protecting 
the surface, as proved by the rock shaft recently sunk out of the 
influence of the Kimberley mine to a depth of one thousand feet, 
where a thickness of four hundred feet of amygdaloidal lava with 
the trappean ecca conglomerate above it represents the prevailing 
rocks of the Vaal, Riet, and Orange rivers for a great distance 
below Hopetown. Incredible as it was deemed at the time, my 
story of the small rounded river stone which fell out of the 
unsealed letter placed in my hands by the post-boy, has since 
1 "Precious Stones and Gems," p. 58. 



4 88 THE DIAMOND MINES OF SOUTH AFRICA 

proved to have been the key that has unlocked the vast under- 
ground wealth of South Africa. 1 

" The story I have now to tell of its birthplace and subse- 
quent history will, I know, appear still more incredible, as fabu- 
lous indeed as was that of Sindbad, the Arabian voyager, who, 
with the talisman and magic lamp of Aladdin the Seer, unlocked 
the caverns of Africa's fairy land, and viewed in prophetic vision 




Irregular Crystallization of Diamonds. 

the vast stores of buried treasures, gold, diamonds, and other 
gems, just as we see them now with our magic electric lamp a 
thousand feet down in the dark recesses of the extinct volcano, 
yielding millions of the purest gems upon earth. 

" How came the diamond there in its hard blue matrix of 
ashes and lava, with its accompanying gems, garnets, rubies, 
sapphires, agates, and other gems, the products of solution and 
1 Geological Magazine, Vol. VI, p. 208, May, 1889. 



FORMATION OF THE DIAMOND 489 

heat ? For a substance to crystallize, its molecules must be free 
to move under polarizing and other metamorphic forces influenc- 
ing crystallization ; but the diamond we know is neither soluble 




Irregular Crystallization of Diamonds. 

nor fusible. It is the element carbon crystallized, and is con- 
sumed by heat. How, then, could it survive as a crystal in the 
crater of a volcano ? 

" The key to solve this mystery was placed in my hand over 
half a century ago, by one of the greatest philosophers of the 



490 THE DIAMOND MINES OF SOUTH AFRICA 

age, whose lectures I had the privilege of attending. But it was 
not until I had examined a diamond mine in South Africa and 
speculated upon the apparently irreconcilable phenomena attend- 
ant upon the origin of the diamond in its matrix, that the prac- 
tical application of Faraday's discovery began to dawn upon me. 
' Hold out your hand,' said he, at the close of the lecture that 
fairly electrified the world of science, as with a loud hiss a snowy 
substance, burning like a coal but in reality intensely cold, 
escaped into the palm of my hand from the strong iron vessel 
in which, with a pressure of fifty atmospheres, he had liquefied 
carbonic acid gas the very gas resulting from the combustion 

of the diamond, consisting of 
one atom of carbon and two 
of oxygen. 

" I have shown that the 
sedimentary beds deposited 
from this vast freshwater lake 
attained a thickness of about 
eight thousand feet. The 
lake itself, therefore, prob- 
ably equalled that depth. (?) 
Now the experiments of 
Wyville Thomson and Car- 

Crystal of Diamond, showing Rounded Edges. penter, made during the 

voyage of the Lightning and the Porcupine, proved that at a 
depth of three to four hundred fathoms, the pressure is equal to 
half a ton on the square inch ; at a mile to one hundred and 
fifty-nine atmospheres, and at seven thousand feet it amounts to 
two hundred atmospheres, or four times the pressure under which 
Faraday liquefied carbonic acid gas, the temperature at such great 
depths being very few degrees above freezing point. In the 
carbonic acid gas generated from the carbonaceous shales by 
heat, and interspersed as gas bubbles in the cavities of the viscid, 
ferruginous amygdaloid, and in the admixture of steam, lava, and 
ashes known as the ' Kimberley Blue ' reduced to the liquid 
state by the enormous pressure in the subaqueous volcano we 




FORMATION OF THE DIAMOND 



491 



have the constituents of the diamond in a form admitting of 
crystallization, and the subsequent absorption of its oxygen by 
the iron always present in its containing walls during long inter- 
mittent periods of volcanic inactivity. There are proofs in the 
Kimberley mine that such alternating periods of activity and 
repose have occurred at long intervals, as shown by the four or 




Twin Crystal Formations. 

five distinct and separate layers of diamonds lining its walls, of 
varying size and quality, known and recognizable by diamond 
buyers." 

In this presentation, which Dr. Atherstone seemingly re- 
garded as conclusive, there is a lack of the clear, logical reason- 
ing which in other discussions has distinguished his views. He 
dogmatically puts the carbonic acid gas evolved from the car- 



492 THE DIAMOND MINES OF SOUTH AFRICA 

bonaceous shales into cavities of the amygdaloidal rock which 
lies outside of the volcanic pipes. Then he reduces this gas by 
enormous pressure to a liquid state, and, having gotten it into a 
form, as he thought, admitting of crystallization, he absorbs the 
oxygen of the carbonic acid by the iron in the containing walls 
of the craters. Now, as a matter of fact, there are no cavities in 
the amygdaloidal rock underlying the shales, for all interstices 
are rilled with silica in the form of agates, or with calcite. Fur- 
thermore, if carbonic acid had been left in the olivine diabase to 
crystallize, then the resultant diamonds would have been enclosed 
in this formation, which is also contrary to fact, for no diamonds 
have ever been found in the amygdaloidal rock. His main con- 
tention, too, is the derival from the shales of the carbon necessary 
for the formation of diamonds. It will be made clear, subse- 
quently in this discussion, that this assumption is not justified. 

The late Henry Carvill Lewis, M.A., F.G.S., Professor of 
Mineralogy in the Academy of Natural Sciences, Philadelphia, 
U.S.A., advanced the proposition that the diamond is the result 
of the intrusion of igneous rocks into and through the carbo- 
naceous shales, and the crystallization of the carbon throughout 
the rocks, as it cools, from hydrocarbon, distilled from the 
shales that had been broken through. 1 

In support of such a theory, it is claimed that the diamonds 
in the various mines or pipes have different characteristics. It 
is quite true that large parcels of diamonds from the various 
mines have distinctive characteristics, and it can be easily told 
from which mine a parcel of diamonds comes ; but it is very 
difficult to tell in which mine a single stone may have been 
found, though each mine has stones in a great measure peculiar 
to itself. Some observers claim that the broken diamonds 
which are extracted are broken during the process of winning 
them. It is admitted that diamonds may be broken in the 
process of mining and the subsequent operations of winning, 
but these cases are exceptional. Fragments of diamonds are 
very frequently found embedded in the blue ground, and there 
1 "Gems and Precious Stones of North America," George F. Kunz. 



FORMATION OF THE DIAMOND 493 

is no doubt in the mind of any one who has had practical 
experience in finding these fragments that they were not crystal- 
lized where they are found. The fact that no diamond has 
ever been found embedded in the shale itself strikes one as 
conclusive proof that Professor Lewis's theory is wrong. 




Diamonds of Regular Forms. 

Again, would not the intrusion of an igneous rock through 
carbonaceous shales have altered these shales in the vicinity of 
the igneous rock? There is, however, no difference that can be 
detected between the shales at the junction of the pipe and at a 
distance of one thousand feet. Moreover, would not the frag- 
ments of shale enclosed in the blue ground have changed, and 
have lost the carbon which they contain, if diamonds were 
formed from them ? One sees no difference between the shale 
which forms the country rock, and the fragments embedded in 
the blue ground. If such a theory as is attributed to Professor 
Lewis by Mr. Kunz had a shadow of foundation, it is dis- 
pelled by the occurrence of diamonds in the Jagersfontein mine 
in the Orange Free State, some eighty miles from Kimberley. 
In this mine there are no carbonaceous shales surrounding the 
diamond-bearing deposits. The pipe, as far as developed, is in 
quartzite, and it is apparent that the shales never existed here, or 



494 THE DIAMOND MINES OF SOUTH AFRICA 

were denuded before the formation of the diamond-bearing pipe. 
If such denudation had taken place after the filling of the pipe 
with a diamond-bearing matrix, the alluvial deposit of the coun- 
try surrounding this mine must contain diamonds, but no such 
discovery of diamonds has been made. 

The Jagersfontein mine is not the only diamond-bearing pipe 
that has produced diamonds without having shale as a country 
rock. Other pipes or veins have been found both in the Free 
State and the Transvaal, which are, however, of little commercial 
value, owing to the small quantities of diamonds found in them, 
but they are most useful in refuting existing theories, if not in 
the determination of the genesis of the diamond. 

An important contribution to this discussion was made by 
Professor Molengraaf, state geologist of the South African 
Republic, in a monograph on the diamonds at Rietfontein in 
the Transvaal. He stated that "the diamond-bearing breccia on 
the farm was of the same nature as the well-known blue ground 
of the Kimberley mines. The geological position of the volcanic 
chimney at Rietfontein was very different from that of the other 
diamond pipes in South Africa. The latter, of course, all oc- 
curred in a higher or lower horizon of the karoo formation, 
whereas the chimney at Rietfontein seemed to occur in the upper 
parts of the Pretoria beds in a system of strata overlying the 
Magaliesberg quartzite. If that position, which was almost cer- 
tain to his mind, was proved to be correctly determined by a 
later and more careful geological survey of the surrounding 
country, this fact would be of high importance in the discussion 
of the genesis of diamonds. Of the different theories regarding 
this genesis he would only mention three principal ones. 

" He would take up first the theory agitated by Messrs. 
Stanislas Meunier, 1 M. Chaper, 2 and in a somewhat modified form 

1 " Composition et origine du sable diamantifere du Du Toils Pan, Afrique 
australe." Comptes rendus de 1' Academic des Sciences de Paris. Vol. 
LXXXIV, No. VI, p. 250. " Examen mineralogique des roches qui accom- 
pagnent le diamant dans les mines du Cap de Bonne Esperance." Bulletins de 
P Academic Royale de Belgique, jd series, Vol. Ill, No. 4. 

2 " Note sur la region diamantifere de 1' Afrique australe." Paris, 1880. 



FORMATION OF THE DIAMOND 495 

lately by Professor Gamier. They denied the igneous origin of 
the blue ground and the diamonds in it, and considered the blue 
ground to be a kind of mud, or peculiar alluvial deposit, which 
had been forced up by a hydrostatic process. That theory, to 
his mind, had already been proved untenable by several eminent 
geologists. The two remaining theories agreed as far as the 
igneous origin of the blue ground. According to one of these, 
the diamonds belong to the primary constituents of the eruptive 
rock itself, and had crystallized at a great depth under very high 
pressure and high temperature, before an eruption of an explo- 
sive character brought the igneous rock to the earth's surface. 

"According to the second theory, which was discussed by 
Mr. Harger at a meeting of the Geological Society of South 
Africa, the diamonds were formed in the blue ground, during 
its ascension, from carbon borrowed from the carbonaceous 
shales through which the eruptive rock forced its way. Now, 
that theory, although rather weak in his opinion, had been main- 
tained, hitherto, mainly because the geological position of the 
known diamond pipes was such that it could be proved, or, at 
least, be accepted as very probable, that the blue ground had 
forced its way through carbonaceous strata. The discovery at 
Rietfontein deprived that theory of its strength. As already 
pointed out, the chimney at Rietfontein was found in the upper 
Pretoria beds. But in the Pretoria beds, as well as in the forma- 
tions underlying these, strata containing any notable quantities of 
carbon were nowhere to be found in the Transvaal ; so that the 
conclusion might safely be drawn that the igneous blue ground, in 
forcing its way from great depths toward the place where it was 
found, could not borrow any carbon from the surrounding strata 
in order to convert it into diamonds. The discovery at Rietfon- 
tein might afford a valuable argument in favor of the formation 
of diamonds as a primary constituent in breccia, or ultrabasic 
magma at great depth, and geologists were entitled to derive 
from it an argument in favor of the following more general 
thesis : ' The elements of carbon, under the conditions of heat 
and pressure ruling at great depth in the interior of the earth, 



496 THE DIAMOND MINES OF SOUTH AFRICA 




Diamonds of Irregular Forms. 



can only exist and crystallize in the modification called diamonds.' 
This thesis was, of course, in perfect harmony with the latest 
scientific discoveries, especially with the famous experiments of 
Moissan." 



FORMATION OF THE DIAMOND 497 

It was the opinion of the late Dr. Stelzner that the diamond 
was crystallized at great depths and came up with the magma or 
matrix. The following liberal translation from a lecture de- 
livered by Dr. Stelzner before the Isis Society in Dresden on 
April 20, 1893, gives the views of this celebrated geologist : 

" Before I give my own opinion, may I be allowed to recall 
three well-known geological facts : first, that various minerals 
which compose many of the eruptive rocks, for instance the 
olivine of certain basalts, contain liquid carbonic acid, and we 
must come to the conclusion that the molten magma under some 
circumstances must have been impregnated with carbonic acid ; 
second, that the blue ground of Kimberley, as already mentioned 
by Lewis, has a known resemblance to many meteorites ; and, 
third, that a modified form of carbon, besides graphite, similar 
to the diamond, has been met with recently in meteorites. 

" If we take these three facts into consideration, and also 
remember that in most of the localities in which diamond-bear- 
ing alluvial deposits appear (Ural, India, Borneo, New South 
Wales, and in the United States), serpentine (especially perido- 
tite) is to be found, we come to the conclusion that the carbon 
of the diamond itself crystallized when this molten mass, rich in 
magnesium silicate, became cool. In support of this opinion we 
find that in some instances diamonds and garnets (pyrope) are 
found together, showing that they have the same origin." 

For the illumination of the problem of the formation of 
diamonds the experiments of Mr. J. B. Hannay of Glasgow, Pro- 
fessor Dewar, and M. Moissan, and later of Sir William Crookes, 
are of the greatest interest to the scientific world. 

The conversion of a diamond into graphite was effected by 
Professor Dewar, publicly, in London, as far back as 1880. Sir 
William Crookes repeated the same experiment in a lecture at 
the Royal Institution of Great Britain, on June n, 1897, by 
placing a diamond in the electric arc where the temperature was 
3600 C., when it was converted into graphite. 

Among the first attempts to make artificial diamonds may 
be mentioned that of Mr. J. B. Hannay of Glasgow, who com- 

2 K 



498 THE DIAMOND MINES OF SOUTH AFRICA 

menced his experiments in 1879, anc ^ a ft er many trials, some of 
which resulted in violent explosions, he is said to have succeeded. 
The method adopted by Mr. Hannay is described as follows: 

" A tube twenty inches long by four inches in diameter was 
bored so as to have an internal diameter of half an inch. In the 
tube was placed a mixture of ninety per cent of rectified bone oil, 
and ten per cent of paraffin spirit, together with four grammes 
(about sixty-two grains) of the metal lithium. The open end of 
the tube was welded air-tight, and the whole mass was heated to 
redness for fourteen hours ; on opening it a great volume of gas 
rushed from the tube, and within was a hard, smooth mass 
adhering to the sides of the tube. It was quite black, and 
appeared to be composed of iron and lithium, but on a closer 
inspection small transparent pieces were found embedded in it. 
The mass was dissolved, and the small transparent pieces proved 
to be ' crystalline carbon,' exactly like diamonds but almost 
microscopical. 

" Out of eighty complex and extensive experiments only 
three succeeded. Violent explosions were frequent, steel tubes 
burst, scattering their fragments around, and furnaces were blown 
up. ' The continued strain on the nerves,' writes Mr. Hannay, 
' watching the temperature of the furnace, and in a state of ten- 
sion in case of an explosion, induces a nervous state which is 
extremely weakening, and when the explosion occurs it some- 
times shakes one so severely that sickness supervenes.' ' 

Sir William Crookes attributes the possibility of making 
artificial diamonds to the facilities afforded by the enormously 
high temperatures which are obtainable in recent years by the 
introduction of electricity. While electricity has, no doubt, 
played an important part- in the scientific researches during the 
last decades of the nineteenth century, Mr. Hannay's experi- 
ments would indicate that it is not absolutely essential to have 
enormous temperatures or pressures to produce artificial dia- 
monds. Still, Sir William Crookes shows that by means of 
these high temperatures substances such as carbon obey the 

1 Glasgow News. 



FORMATION OF THE DIAMOND 499 

common laws which govern other substances, and can be made 
volatile and fusible under certain conditions. He has demon- 
strated that the temperature necessary to volatilize pure carbon 
is about 3600 C., and that it passes into the gaseous state 
without liquefying, and he infers that, if sufficient pressure were 
applied with the high temperature, liquid carbon would be pro- 
duced which upon cooling would crystallize in diamonds. For 
this product the absence of oxygen is absolutely necessary, as 
the carbon would readily unite with it in the form of carbonic 
acid. It is a well-known fact that iron when melted dissolves 
carbon, and while Moissan discovered that other metals effect 
this dissolution, he found that iron was the best solvent. 

Sir William Croolces went through the process of producing 
diamonds before the eyes of his audience, but was only able to 
show them the result of this experiment by reproducing a lantern 
slide of microscopical diamonds which he had made in the same 
way previously, for it takes a fortnight to separate them from 
the iron and other substances in which they are embedded. 
The scientific principle upon which this experiment rests, accord- 
ing to Sir William Crookes, is that molten iron absorbs carbon, 
and as iron increases in volume as it passes from the liquid to 
the solid state, if the outer crust of the iron is suddenly cooled 
and the centre remains in a liquid state, the enormous pressure 
caused by its expanding while cooling affords the two factors neces- 
sary for the crystallization of the diamond heat and pressure. 

Authorities differ somewhat as to the exact moment when 
molten iron expands on cooling, but it is the generally ac- 
cepted theory that expansion takes place at the moment of 
solidification. It is also a well-known fact that shrinkage or 
contraction takes place as the solidified metal cools. It is there- 
fore possible to obtain enormous pressure in the molten centre 
of a casting by the contraction of the outer shell which has been 
rapidly cooled and the expansion of the inner mass just as it 
begins to solidify. 1 

'American Society Mechanical Engineers, Vol. XVIII, pp. 419 and 431. 
American Institute of Mining Engineers, Vol. XVII, 126 and 1015. 



500 THE DIAMOND MINES OF SOUTH AFRICA 

Sir William Crookes says further, that it has been " conclu- 
sively proved that the diamond's genesis must have taken place 
at great depths under enormous pressure. The explosion of 
large diamonds on coming to the surface shows extreme ten- 
sion." According to my own experience, a diamond never 
explodes. Light brown, smoky diamonds often crack on expo- 
sure to the dry air, but they will remain intact if kept in a 
moist place. The cracking is, therefore, more probably the 
result of heat or drying than of tension or inward pressure. It 
is possible, however, that the greater heat to which the diamond 
is exposed when brought to the surface may expand contained 
gases sufficiently to crack the stone. 

Sir William holds the same view of the formation of the dia- 
mond-bearing pipes which I suggested at the time of my visit to 
the Diamond Fields in I885, 1 that these pipes were volcanoes 
which were filled with the mixture which they now contain while 
it was in the form of mud. My reasons for this theory are fully 
set forth upon another page. Continuing in his lecture, Sir Will- 
iam says : " The ash left after burning a diamond invariably con- 
tains iron as its chief constituent, and the most common colors 
of diamonds when not perfectly pellucid show various shades of 
brown and yellow from the palest ' off color ' to almost black. 
These variations accord with the theory that the diamond has 
separated from molten iron." 

I have a collection of diamonds of all colors, see colored 
Plate opposite, and recently made exhaustive tests in order to 
ascertain whether they contained any iron either in the metallic 
or oxidized state. These experiments were made upon a mag- 
netic separating machine, the field magnets of which attracted 
any mineral which contained iron in a metallic or oxidized state. 
Although some of these diamonds had the appearance of being 
coated with iron in some form, and others were colored dark 
brown and deep yellow, they were in no way attracted by the 
magnet even when excited by a strong electrical current. These 

1 Transactions of the American Institute of Mining Engineers (October meeting, 
1886), Vol. XV, pp. 392-417. 



FORMATION OF THE DIAMOND 



501 



experiments do not, perhaps, disprove the existence of iron in 
the diamond, but they do establish the fact that, if iron does exist 
in an oxidized state, the quantity is infinitesimally small. 

One more theory of the deposit of diamonds in the South 
African fields is deserving of special mention, more for the pur- 
pose of showing to what heights of imagination the human mind 
may soar, than for any scientific value it may have. This is an 
assumption that the diamond deposits came from a fall of meteors, 
"a direct gift from heaven," and was first advanced to notice, 
it is said, by Meydenbauer. Such a theory seems highly fan- 
tastic and is the most improbable of all. The occasional inclu- 
sion of black diamonds in meteorites is well attested, but these 
occurrences are very far from accounting for the formation of the 
South African diamond-bearing deposits. " Bizarre as such a 
theory may appear," says Sir William Crookes, " I am bound 
to say there are many circumstances which show that the notion 
of the heavens raining diamonds is not impossible." The 
"Ava" meteorite which fell in 
Hungary in 1846 contained graph- 
ite in cubic crystalline form which 
G. Rose thought was produced by 
the transformation of diamonds. 
Later Weinschenk found trans- 
parent crystals (diamonds) in the 
Ava meteorite. 

Since it became known that 
diamonds (infinitesimally small, it 
is true, but nevertheless diamonds) 
occur in meteorites, a general 
search has been made for the minute crystals in meteorites from 
Australia and Russia, and from Canon Diablo, Arizona, and dia- 
monds and graphite have been found. 1 

From the above facts and from observations which Sir Will- 
iam Crookes made at Kimberley, he concludes that the genesis 
of the diamonds found in the South African mines was by crys- 
1 Sir William Crookes's lecture. 




A Microscopical Diamond (magnified 
zoo times) from a Meteorite from 
Canon Diablo. 



502 THE DIAMOND MINES OF SOUTH AFRICA 

tallization of pure carbon in molten masses of iron which form 
a part of the internal regions of the earth. 

The theory that the diamonds must have crystallized in 
a matrix of iron is not new. That small diamonds have been 
produced in this way there is no doubt, and in the absence of 
further proof to the contrary one might assume that such was 
the origin of the diamond. Iron in the form of magnetite and 
other similar minerals forms a considerable part of the concen- 
trates from the washing machines ; but all proof that these 
minerals, which may have been derived from metallic iron 
by oxidation, were the matrix in which the diamonds originally 
crystallized is wanting. As a matter of fact, I am positive that 
neither the iron nor, as others have asserted, the olivine found 
with the diamonds is the original matrix of the diamond ; and my 
assurance rests upon the fact that no diamonds, however small, 
have ever been found in the iron combination, or in the other 
minerals which accompany them, although these concentrates 
have passed daily under the eyes of hundreds of keen-eyed 
sorters for more than thirty years, and thousands upon thousands 
of tons have been looked over, not once, but at least four times. 
The pieces of the iron minerals and especially of the olivine are 
often very large, quite large enough to contain diamonds weigh- 
ing several carats, which in many cases would have been exposed 
to view had these minerals been the original matrix. We must, 
therefore, look to other sources for the genesis of the diamond. 
I have been of the opinion that diamonds crystallized in very 
much the same way as quartz or other minerals, but under 
peculiar circumstances possibly of pressure and heat. Professor 
Crookes states that diamond crystals are almost invariably per- 
fect on all sides. As a rule this is the case. Quartz crystals 
have been found which have been formed without any attach- 
ment to other substances, that is, with both ends showing pyram- 
idal facets. The same formation may be seen in a great many 
other minerals, and this is usually a characteristic of the diamond, 
but diamonds are found which have been crystallized with some 
portion of the surface resting upon or adhering to some other 



FORMATION OF THE DIAMOND 



503 



substance. The several reproductions of the various forms and 
sizes of diamonds will give the reader some idea of the eccen- 
tricities of these stones. 

The experiments of Herr W. Luzi 1 of Leipsic in the pro- 
duction of artificial figures of corrosion on rough diamonds are 
of exceeding interest in the light which they throw on the crys- 
tallization and the probable matrix and genesis of the diamond. 

Until lately the only appearance of chemical corrosion upon 
the surface of rough diamonds was the regular, triangular, nega- 
tive pyramids, which were pro- 
duced through heating the 
diamonds in the open air, or 
under oxygen flame. Herr 
Luzi has succeeded in pro- 
ducing different and peculiar 
kinds of figures. He discov- 
ered that the breccia from the 
South African diamond mines 
(that is, the matrix or blue 
ground), when in a molten 
condition, possesses the prop- 
erty of absorbing the diamond 
or of changing its shape. 

He describes his experi- 
ment as follows : A small quantity of blue ground was melted 
in a crucible placed in a Fourquinon-Leclerq furnace at a tem- 
perature of 1770 R., which was the highest temperature attain- 
able. A diamond with perfectly smooth natural faces was 
submerged in this molten mass. A further quantity of blue 
ground was then added to the contents of the crucible until it 
was completely filled. A tightly fitting cover was placed on 
the crucible, which was placed in the furnace and again exposed 
for thirty minutes to the greatest heat attainable. When the 
crucible was cooled the diamond was removed and found to be 

1 "Artificial Figures of Corrosion on Rough Diamonds," Berichte der Deut- 
sche n Chemise hen Gese Use haft, 1892. 




Smooth Surface of Diamond dissected by Com- 
bustion. (Magnified loo times.) From a 
photograph by Sir William Crookes. 



5 o 4 THE DIAMOND MINES OF SOUTH AFRICA 

covered with irregular oval and half-round grooves of various 
depths. In one experiment, the diamond was found to be 
deeply eaten away on one side, so that the depression nearly 
penetrated through the stone. 

Diamonds thus magmatically corroded have a similarity, as 
regards the appearance of the corrosion, to hornblende and kin- 
dred materials. A small spot or scar was, at times, found at the 
bottom of a large indent. The diamonds were usually found, 
after the experiments, to be blackened, or covered with a red 
coating, which proved to be oxide of iron. 




Diamonds of Irregular Forms. 

Some of the diamonds showed little black or greenish black 
balls located exactly in the centre of the holes. The formation 
of the balls is doubtless connected with the creation of the 
grooves. These little balls are magnetic, and when treated with 
hydrochloric acid, in which they are only partly soluble, they 
evolve a gas. 

The quantity of these balls was too limited to permit of any 
very exact investigation of their nature. Herr Luzi presumes 
that they are transformed diamond-carbon, i.e. a different modi- 
fication of carbon, which contains either oxide of iron or metallic 
iron reduced out of the oxide. He was, however (owing to the 



FORMATION OF THE DIAMOND 



505 



cost of the material to be experimented upon), unable to deter- 
mine positively what chemical action took place during the time 
the diamonds were 
heated in the com- 
plicated silica flux. 
Some of these partly 
absorbed diamonds, 
upon which Herr 
Luzi experimented, 
are deposited in the 
mineralogical mu- 
seum of the Leipsic 
University. 

Herr Luzi fur- 
ther remarks that 
perhaps other 
molten silica combi- 
nations, or those of 
a similar nature to 
the blue ground, 
may have the same 
power of attacking 
the diamond. 

The knowledge 
that diamonds can 
be absorbed by a sil- 
icate magma makes 
one inclined to in- 
vestigate further the 
genesis of the dia- 
mond, which many 
claim was formed 

under great heat and Two Views of the Face of a Rough Diamond, as seen through 
T - the Microscope. (Magnified 100 times.) 

pressure. It such 

was the genesis of the diamond, Herr Luzi's experiments would 

indicate that the original matrix was not a silica combination 




506 THE DIAMOND MINES OF SOUTH AFRICA 

such as the present blue ground. They tend to prove, rather, 
the theory, which I advanced more than sixteen years ago, that 
the blue ground which contains the diamond owes its formation, 
as it at present exists, more to aqueous than igneous agencies. If 
the diamond is unable to withstand the corroding influence of the 
silica magma at the comparatively low temperature given above, 
how could it possibly have retained its forms of crystallization 
and perfect faces at the far higher temperature and pressure which 
must have existed under the volcanic or igneous theory ? 

It seems a pity that Herr Luzi did not state the exact weight 
of the diamonds upon which he experimented both before and 
after his experiments. The burning or absorption of the dia- 
mond in its matrix would be a strong argument against the 
diamond having been crystallized in situ, or that it came up in its 
present matrix when such matrix was in a molten state. If a 
diamond, subjected in its own matrix or magma in an ordinary 
graphite crucible to a temperature of 1770 R., changes its shape 
and appearance as described by Herr Luzi, could it be expected 
that many diamonds in our mines should be found perfect in 
shape, without a flaw or spot, and with clear, transparent sides, 
so smooth that they have the appearance of having been pol- 
ished ? Nevertheless, such is the appearance of nearly all South 
African diamonds. It would seem from the evidence brought 
forward that only one conclusion is possible, namely, that the 
blue ground in its present state is not the magma of the dia- 
mond. What the original magma or matrix was is unfortu- 
nately far less certain. Some years ago a diamond, weighing 28| 
carats, was found at Kimberley. The external surface of the 
diamond was smooth and crystallized, showing no other mineral 
except the diamond itself. The interior of the diamond was 
white, but not transparent, and, owing to its peculiar appearance, 
the valuator broke the stone in order to satisfy his curiosity. 
The result of the breaking is shown in the full-size illustration 
on page 507. A small perfect octahedral diamond was en- 
closed in the centre of the larger diamond. Nor was this all. 
There were flakes of a white mineral, not diamond, attached to 



FORMATION OF THE DIAMOND 507 

the fragments of the broken diamond. A few grains of these 
were collected and analyzed by Professor Lawn, of the Kim- 
berley School of Mines. In appearance the flakes were white, 
translucent, and crystalline, and about as hard as the steel blade 
of a knife. When heated in a closed tube, moisture was 
given off. The mineral was very slightly effervescent, prob- 
ably due to a trace of carbonate of lime. It fused readily on 
platinum wire to a white bead. 




Diamond bearing a Smaller Crystal in its Centre. 

The mineral was determined to be apophyllite, a silicate of 
lime and potash with 16 per cent of water. If a mineral, which 
is fusible at the ordinary temperature obtained with a blowpipe, 
and which contains 16 per cent of water, was formed at the same 
time the diamond crystallized, it is certain that this did not take 
place under the condition mentioned above, i.e. under enormously 
high temperature. How, then, one may ask, did the apophyllite 
become a part of this stone ? 

Von Tschudi describes a beautiful crystallized Brazilian dia- 



5 o8 THE DIAMOND MINES OF SOUTH AFRICA 

mond, in the centre of which is a little gold leaf. He had the 
information from Dr. Mills Franco, who maintained that there 
was no deception in its being gold. 1 Occurrences of this nature 
tend to veil the genesis of the diamond in still further mystery. 

Professor T. G. Bonney lately obtained specimens from the 
Newlands mines, some forty miles northwest of Kimberley, of a 
coarsely crystalline rock studded with garnets, technically " holo- 
crystalline allied to eclogites," which were embedded, as he says, 
in typical blue ground. In this eclogitic rock he found a num- 
ber of small but perfectly formed diamonds. At a meeting of 
the Royal Society in July, 1899, ne presented his conclusions: 
"The blue ground is not the birthplace, either of the diamond 
or of the garnets, pyroxenes, olivine, and other minerals, more 
or less fragmental, which it incorporates. The diamond is a 
constituent of the eclogite, just as much as a zircon may be a 
constituent of a granite or a syenite. 

" Though the occurrence of diamonds in rocks with a high 
percentage of silica (itacolumite, granite, etc.) has been asserted, the 
statement needs corroboration. This form of crystallized carbon 
hitherto has been found only in meteoric iron (Canon Diablo), 
and has been produced artificially by Moissan and others with 
the same metal as matrix. But in eclogite the silica percentage 
is at least as high as in dolerite ; hence it is difficult to under- 
stand how so small an amount of carbon escaped oxidation. 

" I had always expected that a peridotite (as supposed by 
Professor Lewis), if not a material yet more basic, would prove 
to be the birthplace of the diamond. Can it possibly be a deriv- 
ative mineral, even in the eclogite ? Had it already crystallized 
out of a more basic magma, which, however, was still molten 
when one more acid was injected and the mixture became such 
as to form eclogite ? But I content myself with indicating a 
difficulty and suggesting a possibility ; the fact itself is indispu- 
table : that the diamond occurs, though rather sporadically, as a 
constituent of an eclogite, which rock, according to the ordinary 
rules of inference, would be regarded as its birthplace. 

1 "Travels in South America," by J. J. von Tschudi. 



FORMATION OF THE DIAMOND 509 

" This discovery closes another controversy, viz., that con- 
cerning the nature of the ' hard blue ' of the mines (kimberlite 
of Professor Lewis) in which the diamond is usually found. The 
boulders described in this paper are truly water-worn. The 
idea that they have been rounded by a sort of ' cup and ball ' 
game played by a volcano may be dismissed as practically im- 
possible. Any such process would take a long time, but the 
absence of true scoria implies that the explosive phase was a 
brief one. They resemble stones which have travelled for several 
miles down a mountain torrent, and must have been derived 
from a coarse conglomerate, manufactured by either a strong 
stream or the waves of the sea from fragments obtained from 
more ancient crystalline rocks. . . . 

" The presence of water-worn fragments, large and small, in 
considerable abundance, shows the blue ground to be a true 
breccia, produced by the destruction of various rocks (some of 
them crystalline, others sedimentary, but occasionally including 
water-worn boulders of the former), i.e. a result of shattering 
explosions followed by solfataric action. Hence the name kim- 
berlite must disappear from the list of peridotites, and even from 
petrological literature, unless it be retained for this remarkable 
type of breccia. 

" Boulders, such as we have described, might be expected to 
occur at the base of the sedimentary series, in proximity to a 
crystalline floor. The karoo beds in South Africa . . . are 
underlain in many places by a coarse conglomerate of consider- 
able thickness and great extent, called the Dwyka conglomerate, 
which is supposed to be Permian or Permo-carboniferous in age. 
It crops out from beneath the karoo beds at no great distance 
from the diamond-bearing district and very probably extends 
beneath it. If this deposit has supplied the boulders, the date 
of the genesis of the diamond is carried back, at the very least, 
to Palaeozoic ages, and possibly to a still earlier era in the earth's 
history." 1 

1 Proceedings of the Royal Society, Vol. LXV, pp. 235, 236, July 27, 1899. 
"The Parent Rock of the Diamond in South Africa," Prof. T. G. Bonney. 



510 THE DIAMOND MINES OF SOUTH AFRICA 

I cannot accept the contention that the boulders came from 
any strata through which the pipes have been formed, unless 
these strata lie very deep and below the quartzite. 

The conglomerate which lies between the shale and mela- 
phyre is only a few feet thick, ten to fifteen at most, and does 
not contain large boulders such as are found in the blue ground ; 
besides, the quantity of boulders or conglomerate which could 
have been contained in the area of the mine would not have 
supplied the amount of stones already found in the blue ground. 
These must, therefore, have come up from below with the dia- 
mond-bearing ground. If the boulders came from the Dwyka 
conglomerate, it must lie very deep beneath the surface, for 
nothing of the kind has been found at a depth of over twenty- 
one hundred feet. 

Professor Bonney says above that the statement of the occur- 
rence of diamonds in itacolumite l needs corroboration. There 
is no doubt in my own mind that diamonds in Brazil have been 
found in itacolumite, and the consensus of opinion is that it is 
not the original matrix, but that the diamonds were washed from 
their volcanic origin and became bedded in this sandstone when 
it was being formed. 

I have been frequently asked, " What is your theory of the 
original crystallization of the diamond ? " and the answer has 
always been, " I have none ; for after seventeen years of thought- 
ful study coupled with practical research I find that it is easier 
to ' drive a coach and four ' through most theories which have 
been propounded than to suggest one which would be based 
upon any more unassailable data." All that can be said is that 
in some unknown manner carbon, which existed down deep in 
the internal regions of the earth, was changed from its black and 
uninviting appearance to the most beautiful gem which ever saw 
the light of day. 

1 Brittle quartz sandstone of slaty (schistose) character. Heusser. 



CHAPTER XVII 



THE DIAMOND MARKET 




N preceding chapters the extraction of the blue 
ground and the winning of the precious stones 
have been fully described. It remains to trace 
the handling of the diamonds from this point 
until they reach the hands of the jewellers and 
are spread broadcast in glittering array over the 
face of the world, or applied to uses less showy than adornment. 
After the diamonds are separated and collected at the Pul- 
sator, they are cleaned and sent under guard to the diamond 
office, which is in the general offices of the Company. Here 
the crystals are boiled in a mixture of nitric and sulphuric acid 
to remove any particles of earth which may adhere to them. 
They are then thoroughly rinsed with clear water to get rid of 
the acids, and finally washed in alcohol and spread out on tables 
to dry. The alcohol seems to clean the diamonds and leaves 
them brighter than when water alone is used. 

The daily productions of diamonds are put away in parcels 
until there is an accumulation of about 50,000 carats of De 
Beers and Kimberley diamonds. The diamonds from these 
two mines are mixed and are known locally as " pool goods." 
When the requisite quantity is at hand, the mixed stones are 
screened to grade the sizes, after first taking out the larger 
diamonds by hand. They are then ready to pass to the hands 
of the sorters, who separate and classify them for accurate valua- 
tion. The chief classifications in use are 

i. Close goods. 2. Spotted stones. 3. Rejection cleavage. 
4. Fine cleavage. 5. Light-brown cleavage. 6. Ordinary and 
rejection cleavage. 7. Flats. 8. Maacles. 9. Rubbish. 10. Boart. 

5" 



5 i2 THE DIAMOND MINES OF SOUTH AFRICA 

" Close goods " are pure, well-shaped stones ; " spotted 
stones " are crystals slightly spotted ; and " rejection " stones 
seriously depreciated by spots. Broken stones are grouped 
under the head of " cleavage." Flat crystals formed by the 
distortion of octahedra are classed as flats, and flat triangular 
crystals, which are in reality twin stones, are marked as maacles. 
" Rubbish " is the refuse, ranking a little better than the lowest 
grade of all, ordinary " boart " the material used for polishing 
purposes. Round or shot boart is found in the mines at 




Diamond Sorters and Valuators, through whose hands ,4,000,000 worth of Diamonds 

pass annually. 

Kimberley and is very valuable for use in diamond drills since 
the Brazilian carbonado has become so scarce. Well-formed 
shot boart, averaging about the size of peas, sells readily for 6 
a carat. 

After this separation has been made, the first eight classes are 
each further subdivided according to their shades of color. The 
scale is given below in descending order of purity 

Blue White, First Cape, Second Cape, First Bye, Second 
Bye, Off Color, Light Yellow, Yellow. 

Only the first grade, or close goods, are carefully distin- 
guished by separation of all eight shades. For other classes a 



THE DIAMOND MARKET 513 

smaller number of shade divisions is noted. It may be per- 
ceived that the minute distinctions of this separation can only be 
made by the trained eyes of experts. No magnifying glasses 
are used by the sorters, all being able to make the distinctions 
with the naked eye. Ten sorters are employed, all Europeans, 
two women and eight men. To replace any who leave, ap- 
prentices are trained to the work at Kimberley. The sorters 
determine the quality of diamonds with notable accuracy and 
speed. 

De Beers mine is noted for yielding an exceptionally large 
percentage of ordinary " yellows," a very small percentage of 
very " dark yellows," a limited number of brilliant " silver 
Capes," and considerable " light-brown cleavage " of a delicate 
shade. The very " dark yellows " are ranked as " fancies " and 
highly valued, and the "silver Capes" are also rated highly, as 
they have great lustre when cut as brilliants, but absolutely 
white or colorless stones are rarely found in this mine. 

Kimberley mine yields a fair proportion of " white crystals," 
a good percentage of " white cleavage," and quite a remarkable 
percentage of large " maacles." It also produces a fairly large 
proportion of " yellows," generally somewhat lighter in color 
than those from De Beers. 

Dutoitspan mine yields some very fine blue-white stones, 
" silver Capes " and ordinary " white " stones and " cleavage " 
of comparatively fine quality, together with large " yellows," 
showing an exceptional proportion of large stones, and a com- 
paratively small percentage of very minute crystals. 

Bultfontein's product is very largely composed of white 
stones, but many of these are spotted more or less ; its diamonds 
are also comparatively small, usually ranging from two to three 
carats downwards. 

The diamonds from Premier mine are mostly octahedron 
crystals, or fragments of these, with a large percentage of rub- 
bish and boart. Beautiful, deep-orange colored diamonds are 
frequently found, and blue-white stones are not uncommon. 

When the sorting has been completed, the diamonds are 

2 L 



THE DIAMOND MINES OF SOUTH AFRICA 

placed in little heaps on a long table covered with white paper. 
In all cases, except in small sizes and boart, where the weight 
and value only are recorded, the number of diamonds in each 
heap and their average weights and values are carefully recorded 
in a book kept for that purpose. This exhibit was previously 
made also for the benefit of buyers calling at the diamond office, 
who could thus readily value the stones ; but of late years the 
entire product has been sold to a syndicate composed of the 
leading diamond merchants of Holborn Viaduct and Hatton 
Garden, London. The careful sorting and arrangement are 
nevertheless continued in order to determine precisely what the 
relative quality and value of the diamonds are in passing from 
level to level as the mine grows deeper. The buyers know the 
exact value of every shipment they make, and the De Beers 
Company must also be informed of any changes for better 
or for worse in the value of its production, so as to take ad- 
vantage of them in the former case, or make allowances to 
the syndicate upon the renewal of the contract, in case the 
quality should become poorer. These are perhaps remote con- 
ditions, for, up to the present time, the average monthly or 
annual production of diamonds has been remarkably regular in 
quality. 

For the safe-keeping of the gems in the Company's office 
there is a strong room or vault, built of very thick concrete 
walls, which are fire and burglar proof. The door of the vault 
is secured by several bank locks of the latest and best design. 
The keys fitting these locks are kept by several officers in the 
secretary's department of the Company, who must all be present 
at the opening and closing of the strong room. Inside the 
strong room are burglar-proof safes, with doors also secured by 
several locks, which can only be opened by two or more persons 
having separate keys. In addition to these safeguards, the 
strong room is protected by the application of an electric alarm 
system. Two armed guards are on duty at the offices at night, 
and connections are made by which they can signal for help 
should an attempt be made to break into the building. Even 



THE DIAMOND MARKET 



if both men should be overpowered before they could give 
a signal, no robbery could be effected ; for, as soon as they 
should cease to send test signal reports at regular intervals, an 
armed force would soon arrive on the ground and frustrate any 
attempted burglary. Under existing conditions for the sale of 
diamonds only a small quantity of precious stones are kept at 
the diamond office ; but, in former years, the quantity, at times, 
has been very large and the most stringent precautions were 




A De Beers Group. 

necessary. It may be noted further that adequate measures 
have been taken also to protect the office from assault in the 
daytime. 

Of late years, with improved methods of working, a larger 
percentage of small diamonds has been recovered from the blue 
ground. In order to have an average quantity of these in 
each parcel made up for the buyers, a fixed percentage of small 
stones is included in the parcel. If there is any surplus, it 
is valued in the ordinary way and sold to the buyers at a valu- 
ation agreed upon between the seller and buyer. After the 



5i6 THE DIAMOND MINES OF SOUTH AFRICA 

diamonds are sorted, they are put into square tin boxes, fitting 
into tin cases like despatch boxes, which have tightly fitting, 
locked covers. A despatch box will contain about forty tin 
boxes. 

All De Beers diamonds are delivered to the buyers at the 
diamond office of the Company and paid for at once in cash or 
in bills on London, as the Company may prefer. After delivery 
to the buyers the diamonds are sorted over again for the London 
market, which desires a classification of the stones for different 
purposes than valuation simply. 

They are reasserted according to quality into from 350 to 400 
different parcels. Each parcel is put into specially made papers 
bearing on their face a description of their contents. Then 
these parcels are packed in tin boxes which are securely wrapped 
in cloth-lined packing paper, carefully sealed and delivered to 
the post-office, which forwards them to Europe as registered 
mail. All diamonds so forwarded are insured with insurance 
companies in Europe. 

Classification is made into 

Pure goods "i 

Brown goods I Completely formed 
Spotted goods or crystallized stones. 

Flat-shaped goods J 

Pure cleavage "1 

. . ' Broken crystals 

bpotted cleavage > .. 

_* or split stones. 

Brown cleavage J 

XT A/I i f Flat, triangular crystals, 

Naats or Maacles { . .. . J 

[_ m reality twin stones. 

fUncuttable diamonds used mostly 
for splitting and polishing more 
perfect crystals. 

Most of the above classifications, except rejections and boart, 
are subdivided into six or seven colors, and each color is again 
subdivided into eight, ten, or twelve sizes. 



THE DIAMOND MARKET 



Si? 



THE PERCENTAGES OF DIAMONDS IN THE VARIOUS CLASSES 





DE BEERS AND 

KlMBEKLEV 

MINES 
POOL GOODS. 


PREMIER 

MINE. 


JAGERSFONTEIN MINE. 


Pure 
Stones 

Spoiled 
Cleavag 
one ef 
Chips, i 
under 
Chips, s 
under 
Maacles 
pure a 
Rejectio 
descri 
Boart, d 
cutlim 


' Close goods, stones over 


4-i% 
1.6% 

3-1% 
1.6% 

8.1% 
38-8% 
I. % 

12.6% 

4-3% 
17-5% 
7-3% 


8.7% all sizes 
4- % 

4-2% 

6.6% 

\ including 
28 - 2 imaacles 

1.2% 

5-5% 

16.2% 
254% 


( all sizes, and includ- 
ii.7%-j ing slightly colored 
I and brown stones 

204% f a " Sizes and de - 
( scnptions 

39- % 
10.6% 

4-3% 
3-5% 
10.5% 


Irregular shapes of all 


Melee o( all sizes under 


,. Brown stones of all sizes. 
Stones of all sizes . . . 
ss, pure and spotted, over 
rat ... 


>ure pieces of all sizes 
one carat 


potted pieces of all sizes 


(Naals) and Flat Stones, 
nd spotted, all sizes . . 
ns, lowest quality of above 
ptions 


iamonds not suitable for 

r 




100. % 


too. % 


100. % 



When the diamonds arrive in London, they are again re- 
assorted for sale, i.e. in the manner that will best suit the customs 
and requirements of the trade. The London importers sell 
(a) to merchants of rough diamonds, who again resell the goods 
in their rough state, (b) to merchants of brilliants who get their 
purchases cut and polished for sale, (c) to actual manufacturers 
who, buying for their own account, cut and polish the goods and 
then resell with profit as compared to the manufacturer who 
works for a fixed cutting charge. 

It is of interest to compare the present elaborate method of 
assorting and valuing with that obtaining in the eighteenth cen- 
tury and previously in the European market. It was the custom 
then to forward diamonds from India in " bulces " or parcels 
neatly wrapped in muslin and sealed by the sellers. The largest 
stones were never offered for sale, but reserved by the native 
owners, as David Jeffries observes, to aggrandize their families. 
He states further that "the head of the family has a small shal- 



5 i8 THE DIAMOND MINES OF SOUTH AFRICA 

low hole drilled in the surface of the stone, and when he dies 
the next chief does the same, and so from one to another, and 
the more of these holes a stone has the higher it is in esteem, 
although such holes may prejudice it if it were to be manufac- 
tured ; but as that is never intended, they do not regard such 
prejudice ; and these stones are never parted with, let what will 
happen, and if they foresee any ruin to the family ... in such 
cases they bury these stones, so that they never appear again." 
The other stones, comprising the small and middle size and some 
of the large ones, were put in the parcels for sale unassorted and 




The Officials who manage the Benefit Society, De Beers Consolidated Mines Limited. 

" valued by the lump, as they weigh one with another, by the 
rule." In the European markets such parcels were generally 
bought, he states, " by the invoice, that is before they are opened, 
it being always supposed they contain the value which they were 
sold for in India; and the buyer here gives the merchant such a 
profit as contents him. The diamonds being thus bought, the 
buyer opens the parcel, separates them, and then values them 
separately as his judgment directs ; making to himself likewise 
such a profit upon the whole parcel as he thinks proper." 
This expert jeweller notes with regret that at the time of his 

1 "A Treatise on Diamonds and Pearls," David Jeffries, London, 1751, 
pp. 1 1 8, 119. 



THE DIAMOND MARKET 




520 THE DIAMOND MINES OF SOUTH AFRICA 

writing, in the middle of the eighteenth century, there was no 
uniform standard of valuation and that the purchase of large 
stones in particular was essentially a gambling speculation. In 
the East Indian market there was a persistent effort to maintain 
fixed prices, and there was comparatively little fluctuation in the 
market rates of the East Indian stones, but the diamonds of 
Brazil were thrown irregularly on the market, so that the supply 
ranged from a dearth to a glut, and the prices were so greatly 
fluctuating that any investment in these stones was extra hazard- 
ous. Mr. Jeffries marked clearly the disastrous consequences 
of greatly varying products and prices in the marketing of 
precious stones. He reached the conclusion " that to maintain 
as invariable a price of these jewels [diamonds] as is possible 
must be of the greatest utility to the public," and gave high 
praise to the owners of East Indian diamond fields and diamond 
merchants because they did not flood the market regardless of 
the diamond, like the Brazilian producers. He notes a shift of 
fully 33 per cent in the market rate of diamonds in a single year. 
In 1733 the value of Brazil diamonds fell to a point below 20 
shillings per carat for rough diamonds, and within 20 years ran 
up to more than treble this price. 

One of the simplest and oldest divisions in grading and in 
the measure of values of diamonds and other precious stones is 
in accordance with their weight. The transmitted measure of 
weight is the carat, derived from the Greek Kepdnov, the fruit 
of a variety of acacia, whose remarkably uniform seeds served as 
convenient measures of value of diamonds and other precious 
stones, and is equivalent to 4 grains avoirdupois or 3.174 grains 
troy weight. In market quotations from year to year and in 
contracts for sorted diamonds the valuation is expressed in a 
stated price per carat. Von Tschudi states that the word carat 
is derived from kaura, an African creeping plant, whose red 
seeds specked with black were used for weighing gold in Africa 
and diamonds in India. On the supposition that 1 may be 
reckoned the general or average price of a rough diamond of 
one carat weight, Mr. Jeffries gives two methods or formulas 



THE DIAMOND MARKET 521 

for computing the values of " wrought," or as we would say 
" cut," diamonds. First, the weight of the cut stone should be 
doubled, to offset the loss of one-half in working ; then this 
figure or figures should be squared, and the product multiplied 
by the price per carat. Thus a cut stone weighing one carat 
would be valued by multiplying 2 by 2 by 2, or at ^8, and a 
stone weighing 5 carats by multiplying 10 by 10 by 2, or at 
^200. By the second method, the calculation is made on the 
basis of the valuation of a cut stone weighing one carat, at _8, 
as before determined. Then to find the value of a stone of any 
given number of carats, multiply the number by 8, and the multi- 
plicand will be the estimated value of every carat in the stone. 
The total value may then be reached by multiplying the number 
of carats by this multiplicand. For example, if a given stone 
weighs 5^ carats, the value of every carat in the stone will be 
found by multiplying by 8 to be ^\. Then multiply ^41 by 
5^ and the result will be ^210 is. 6d. ; the estimated value of 
a cut stone weighing 5^ carats. It was the expectation of Mr. 
Jeffries that the general adoption of his method of valuation 
would go far to fix the price of diamonds, and it did prevail for 
more than a century before falling into disuse. 

Production of Diamonds 

De Beers Consolidated Mines. During the eleven years end- 
ing June 30, 1899, the yield of De Beers Consolidated Mines 
has been 24,476,000 carats of diamonds in round figures, which 
would measure about 72 cubic feet, showing an average of some- 
thing more than 2,200,000 carats annually. Compared with 
this product, the production of the other diamond fields of the 
world, with the exception of Jagersfontein, is comparatively un- 
important, not exceeding 5 per cent of the total. 

The Orange River Colony. The principal diamond-producing 
mine in this colony is Jagersfontein, which has averaged about 
250,000 carats annually for several years past. The Jagers- 
fontein mine is controlled by the syndicate which has for many 
years purchased the total production of De Beers Company. 



522 THE DIAMOND MINES OF SOUTH AFRICA 

There are a few other diamond mines in the Orange River 
Colony, but the yield of diamonds from all of them combined 
is small in comparison with Jagersfontein. The total output of 
diamonds from Jagersfontein up to March, 1901, was 2,168,399^ 
carats, valued at ^3,923,940. 



PUI h.,,. 

-*', men- 




Photograph of a Book with the Leaves cut out in the Centre, used by an Illicit Diamond 
Dealer to send Diamonds through the Mails to England. 

Transvaal. There are alluvial diggings along the banks of 
the Vaal River a few miles above the river diggings in Griqua- 
land West, Cape Colony. The town of Christiana is situated 
near these diggings, lying just outside the jurisdiction of the 
Cape Colony and the late Orange Free State (now Orange River 
Colony), in which Colonies the Diamond Trade Act, which for- 
bids dealing in rough diamonds except by licensed dealers, is in 



THE DIAMOND MARKET 523 

force. A large illicit trade has been carried on at Christiana for 
many years in diamonds stolen in Kimberley and the river dig- 
gings in the Cape Colony. A few years ago the Government of 
the late South African Republic passed certain laws in reference 
to the registration of diamonds, but these laws were not stringent 
enough to stop the illicit traffic. Diamonds have also been 
found at Rietfontein, near Pretoria, but up to the present time 
the total yield has been very small. A few years ago there was 
a remarkable occurrence of diamonds in the conglomerate gold 
ores from the mines at Klerksdorp, when several green diamonds 
were found in the battery box. As the conglomerate is a sedi- 
mentary formation, the diamonds must have been washed into 
it from some crater in a similar manner to the depositing of 
diamonds in the itacolumite of Brazil. 

Outside of South Africa the diamond fields of any deter- 
mined value are in Brazil, India, New South Wales, and Borneo. 

Brazil. There was a revival of the diamond-mining industry 
to some extent in the Brazilian fields, owing to the diminution 
of the South African product by the Transvaal War. The State 
places a duty of 16 per cent on the valuation of all diamonds 
produced, and there is in addition a tax of i per cent demanded 
by the municipalities. Owing to the tax evasion, it is difficult 
to determine the total annual product. The value of exports 
from Minas Geraes during the first half of 1900 was reported 
at 250,000 milreis, $140,000. 

Mr. A. de Jaeger has estimated the total production of 
Brazilian stones from the time of the discovery of the diamond 
fields at 12,000,000 carats, valued roundly at $100,000,000. 
It is stated, however, in "The Mineral Industry," presenting 
probably the best extant record, that the best available statistics 
show that the total output of Brazil, up to and including 1898, 
was 13,105,000 carats. 1 

Dr. Le Neve Foster, one of his Majesty's inspectors of 
mines, in his Annual Report on Mines for 1899, says: "Com- 
pared with the output of Kimberley, the total production of 
'"The Mineral Industry," 1899, p. 222. 



524 THE DIAMOND MINES OF SOUTH AFRICA 

diamonds in Brazil for the year, estimated at 40,000 carats, is 
at present insignificant. . . . The most important diamond dis- 
tricts in Brazil are Diamantina, Grao Mogul, Chapada Diaman- 
tina, Bagagem, Goyaz, and Matto Grosso." 

India. In the same report the quantity of diamonds pro- 
duced in India for 1898 is given at 170 carats, valued at 10,873 
rupees, and for 1899, 124 carats, valued at Son rupees. 

New South Wales. The existence of diamonds in New South 
Wales was made known as early as 1859, by Rev. B. W. Clarke, 
who received in that year several specimens from the Macqua- 
rie River, Burrendong, and Pyramul and Calabash Creeks. It 
was not, however, until the rush for the gold diggings, seven or 
eight years later, that any considerable number of diamonds was 
found, when the gold digging along the Cudgegong River, about 
nineteen miles northwest of Mudgee, brought to light diamonds 
in an old river drift, generally covered with a layer of basalt. 

The diamonds were sparsely distributed through the gravel, 
and were usually small, the largest of the stones, a colorless 
octahedron, weighing only |- carats. Later, other diamond 
fields were opened near Bingera, on the river Hoclon, and in 
the tin-mining districts near Inverell. The diamonds occur in 
alluvial gravel wash in the beds of ancient rivers. This gravel 
carries tin ore or gold in places, and usually one or both of 
these are won with the diamonds. These ancient river channels 
resemble those in California, in which diamonds were occasionally 
found with the gold. Many of these rivers lie buried beneath 
lava hundreds of feet thick, and the diamonds are won by driv- 
ing long tunnels and drifting out the gravel lying on the bed rock. 

Dr. C. Le Neve Foster gives the production in New South 
Wales for 1898 as 16,493 carats, valued at ^"6060, and for 
1899, 25,874 carats, valued at .10,350. These figures give an 
average value per carat of seven shillings and four pence and 
eight shillings respectively, as compared with forty shillings per 
carat for De Beers and Kimberley mines. 

Borneo. The estimated production of diamonds in Western 
Borneo was 1190 carats for 1897, anc ^ I 95 carats for 1898. 



THE DIAMOND MARKET 525 

British Guiana. Some attention has been drawn of late to 
the reported diamantiferous deposits in British Guiana. It is 
stated that there was a shipment of 282 specimens from this 
field to London early in 1900, and, later in the year, 400 small 
stones were brought to Georgetown. The location of the de- 
posits is reported to be on the Mazaruni River, about 250 miles 
from its mouth. The diamonds have been found in an alluvial 
formation, consisting of sandy clay mixed with pebbles and frag- 
ments of ironstone, quartz, and felsite. 1 

Importation of Diamonds 

In the importation of diamonds the United States leads, and 
England, Germany, France, and Italy follow in the order named. 
The increase in the demand of the United States has been ex- 
traordinary, showing an advance of fully 2000 per cent in the 
last fifty years. In 1899 the valuation of the total import of 
precious stones was $17,208,531. In 1900 there was a falling 
oft" of about 13,850,000 owing to the interruption in the supply, 
but the records of the year 1901 indicate a probable importation 
exceeding $20,000,000, the total for the first two months of the 
year reaching $3,870,359.31, an increase of $2,674,787.88 over 
the import of the corresponding months in 1900. The importa- 
tion is a close measure of the total sale, as the production of 
precious stones in the United States only reached a valuation 
of $185,770 in 1899, and this was larger than in any previous 
year. Nearly five-sixths of this native product is made up of 
sapphires and turquoises. 

Rubies, emeralds, sapphires, and pearls are the gems most 
commonly used in settings in combination with diamonds. It is 
estimated by Mr. George F. Kunz, of New York, an expert of 
international reputation, that the value of the diamonds imported 
into the United States is approximately 75 per cent of the valua- 
tion of all precious stories and pearls imported, and it is judged that 
this consumption fairly represented the percentage in other coun- 
tries. The changes in settings from year to year and even from 
decade to decade are not very pronounced. The resetting of 
1 " The Mineral Industry," 1900. 



526 THE DIAMOND MINES OF SOUTH AFRICA 

stones is an appreciable fraction of the jeweller's business, but in- 
considerable in comparison with the setting of newly cut stones. 

The World's Stock 

Diamonds are so highly prized and so imperishable that the 
amount of these gems in existence to-day may almost be reck- 
oned as the total of the world's production, ranging in value 
through hundreds of millions of dollars. Mr. Kunz does not 
estimate a loss of 5 per cent in a hundred years, and the South 
African Diamond Fields alone have contributed over $400,000,- 
ooo, or ; 8 0,000,000, in value to the world's stock. Yet the 
demand advances apace with the world's growth in wealth, and 
no diversion of the world's fancy is apparent. The plunder of 
Delhi by Nadir Shah in 1739 has been estimated at $300,000,- 
ooo, 1 and a great share of this was precious stones. There may 
never again be such a collection in the hands of any monarch or 
nabob as the store amassed by the Great Moguls, but the crown 
jewels and private treasures of the leading courts of Europe 
to-day are of immense value and are growing greater. 

The crown jewels of France were estimated at $6,000,000 
(; i, 200,000) more than a hundred years ago, and even this great 
amount is far exceeded by the value of the Russian crown 
jewels. The crown of Ivan Alexiowitch contained 88 1 brilliants, 
the Empress Catherine had 2536 brilliants in her crown, and 
the purchases of succeeding Czars have been enormous. At the 
London Industrial Exposition in 1851 a firm of Russian jewellers 
exhibited a superb diadem on which were mounted 1 1 beautiful 
opals, 67 rubies, 1811 brilliants, and 1712 rose-cut diamonds. 2 

The British crown jewels do not equal the Russian in num- 
ber or value, though there are other magnificent gems among 
them besides the Koh-i-nur, whose romantic story is told in a 
former chapter. The crown specially made for the coronation 
of the late Queen Victoria, in 1838, was regarded as a superb 
showing of the art of the leading jewellers of London as well as 

1 "Great Diamonds of the World," Streeter. 

2 "A Popular Treatise on Gems," Feuchtwanger, New York, 1867. 



THE DIAMOND MARKET 527 

of the gems displayed. It is fashioned of hoops of silver en- 
closing a cap of deep blue velvet. Precious stones completely 
encase the hoops, which are surmounted by a ball covered with 
diamonds and bearing a Maltese cross of brilliants, with a splen- 
did sapphire as the central jewel. The rim of the crown is 
clustered with brilliants and Maltese crosses. On the cross at the 
front of the crown is set the magnificent heart-shaped ruby, which 
was worn by Edward, the Black Prince, and beneath this ruby in 
a circular rim is an oblong sapphire of extraordinary size and 
beauty. Clusters of drop pearls add to the resplendent effect of 
the massing of the diamonds, emeralds, rubies, and sapphires. 

The exquisite beauty of the jewels of Queen Isabella of 
Spain has been particularly noted. At the London Exhibition 
in 1851 two sets of her jewels were shown. One consisted of 
a diamond necklace, in the form of a ribbon, interlaced with 
foliage of emeralds. Brilliants were arranged also to form a 
bouquet of lilies with emerald leaves, encircled with ribbons of 
brilliants and pendants of pearls. A ribbon of brilliants, inter- 
laced with emeralds, formed a bracelet, and the crown of this set 
was of the like combination of gems, with aiguillettes of flowers 
whose stamens were pearls. The second set of jewels was made 
up entirely of diamonds and sapphires of the finest quality and 
most artfully matched. 

It is scarcely to be expected that any private collections of 
gems should rival in extent the treasures of sovereigns, whose 
crown jewels may be the display of centuries of accumulation, 
but some of the noble families of Europe and other wealthy 
owners have gems that any monarch in the world might covet, 
and there are a considerable number of collections ranging in 
value over a million dollars. In the United States it is esti- 
mated that there are at least half a dozen such collections, one 
of which contains a necklace valued at $^io,ooo. 1 At every 
leading court reception, or grand ball or opera, the display of 
jewels may be measured in millions of dollars, and the diffusion 
of gems is constantly spreading with the extension of wealth. 

1 George F. Kunz. 



CHAPTER XVIII 



CUTTING AND POLISHING 




T has been shown in the opening chapter of this 
work that fancy has still, and probably must 
forever have, a free range for its surmise when 
and how the first diamond crystal was picked 
from the river-shore wash of the Indo-Gangetic 
plain. Equally vague and conjectural must be 
any effort to fix the period when a rough or natural diamond was 
first artificially ground or polished. It is only certain that some 
rude polishing, at least, was essential to the revelation of any 
notable beauty in the diamonds of India ; for the surface of 
these crystals is covered with a grayish white film or incrustation, 
veiling their refulgence so completely that the rough stones 
are scarcely more ornamental than common quartz pebbles. 

It was in view of this obscuring that the apostle of deport- 
ment, the Earl of Chesterfield, wrote to his son : " Manners 
must adorn knowledge and smooth its way through the world. 
Like a great rough diamond, it may do very well in a closet 
by way of curiosity and also for its intrinsic value." A con- 
temporary of this high authority, Dr. Samuel Johnson, was able 
to controvert this dictum by demonstrating that knowledge can 
rise from obscurity without any adornment of manners, but 
polish is indispensable to the revelation of the latent beauties 
of the rough diamond. 

Indian tradition runs back romantically five thousand years 
to the first gleam of the Koh-i-nur or "Mountain of Light" 
in the serpench of a chief who fell in the great battle described 

1 Letters of the Earl of Chesterfield, July I, 1748. 
528 



CUTTING AND POLISHING 



529 



in the epic poem " Mahabharata " ; ' but nothing more solid 
than tradition sustains this tale. If it were true, it would 
demonstrate incontestably a very ancient proficiency in the art 
of grinding and polishing a rough Indian diamond, as the figure 
of the Koh-i-nur on page i shows, illustrating the appearance of 
this famous gem before it was recut by modern lapidary art to 




Examination of the Diamonds before and after Cutting. 

hold the foremost place in the jewels of the British crown. 2 The 
Italian, Augusto Costellani, is the mouthpiece of another tradi- 
tion, little firmer than a floating pipe-bubble, that a certain King 
Carna of India, who lived some three thousand years before 
the Christian era, possessed a diamond whose natural planes or 

1 "Indian Epic Poetry," Sir Monier-Monier Williams, 1863. 

2 "A Popular Treatise on Gems," Dr. Lewis Feuchtvvanger, 1867, Plate 
VIII, No. i 5 and No. i 53, 

2 M 



530 THE DIAMOND MINES OF SOUTH AFRICA 

facets were polished ; but what the good king did with his 
sparkling treasure, or where it has wandered, is unfortunately 
left to the drift of fancy. 

It has been shown that the earliest known catalogues of 
gems do not include the diamond, and that the references to it 
in the Hebrew Scriptures and other writings before the Christian 
era are far from decisive, in view of the likelihood that the 
white sapphire was the ancient adamas. 1 The failure to bring 
to light any diamond in the exhumation of ancient gems is 
further significant. 2 If it be true that a genuine diamond, bear- 
ing the engraved head of the philosopher Posidonius, exists in 
the collection of the Duke of Bedford, as reported by Streeter, 8 
this is a solitary instance, so far as is known, of the application 
of engraving to this adamantine surface at a date probably prior 
to the birth of Christ, for Posidonius was a Tyrian Greek, living 
in the second and first centuries B.C. 4 

It is, however, highly probable that the genuine diamond 
crystals were discovered in India hundreds, if not thousands, of 
years before the Christian era, and partially polished, at least, in 
the primitive method of rubbing or striking the planes of one 
crystal against the other, or even by laborious friction with grit- 
stone by hand or a grinding wheel. 

It is certain that revolving stones or metallic wheels for grind- 
ing gems were in use in remote antiquity, perhaps two thousand 
years or more before the Christian era. From the softer stones, 
carnelian, onyx, and jasper, the ancient workmen advanced to 
harder gems, preparing their face first chiefly by a smooth pol- 
ish for the sculptors of cameos and intaglios. 5 Their mode of 

1 "Precious Stones noted in the Sacred Scriptures," R. Hindmarsh, 1851. 
"Precious Stones and Gems," Edwin William Streeter, 1880. 

2 The Story of the Nations, " Phoenicia," George Rawlinson, M.A., 1894. 
"Ancient Mineralogy," N. F. Moore, 1834. 

3 " Precious Stones and Gems," Streeter, p. 46. 

4 The Story of the Nations, " Phoenicia," George Rawlinson, M.A., 1894. 
"Ancient Mineralogy," N. F. Moore, 1834. 

5 " A Treatise on the Ancient Method of Engraving Precious Stones," Lauren- 
tius Natter, London, 1754. "A Treatise on Diamonds and Precious Stones," 
John Mawe, 1813. 



CUTTING AND POLISHING 531 

working was very simple, as Feuchtwanger notes. 1 The pol- 
ishers prepared the stones on a plate by means of the powder of 
harder stones, either round, oval, flat, or in shield form, accord- 
ing to the designed subjects, and the sculptors cut the engraving 
with iron tools or diamond splinters mounted in iron. 

The Egyptians taught the art of carving to the Phoenicians, 
Etrurians, and Greeks. The Indians and Persians learned to 
carve and polish gems perhaps as early as the Egyptians. Repre- 
sentations of the adored beetles or scarabs were the earliest known 
Egyptian engravings, while the Persians engraved chiefly mytho- 
logical animals or figures of their priests. Cabalistic devices and 
Arabic letters on gems formed the doubly precious "talismans," 
and even without talismanic lettering, marvellous or supernatural 
origin and powers were attributed by current superstition to all the 
notable gems. 3 Alexander's seal typified the sovereignty trans- 
ferred to his vicegerent, Perdiccas. Augustus Caesar cherished 
his seal engraved with a sphinx as a token of his divine authority. 2 

In the carving of cameos, precious stones with layers and 
veins were employed with great skill, bringing out contrasted 
effects, as where a face is shown in one color and the hair and 
dress of a figure in different colors. Sometimes certain colors 
were made typical. Thus Bacchus was carved in amethyst, the 
color of wine, while Neptune or nymphs of the sea were cut in 
aquamarine. 8 

Such surface polishing and engraving antedated, however, 
very far any grinding or faceting of the harder gems, and the 
intractable diamond especially, for uses of ornament. Pliny 
writes, "The polished hexahedral Indian diamond thins to a 
point." 4 As the crystallization of the diamond is much more 

1 "A Popular Treatise on Gems," Feuchtwanger, 1867. 

2 De duodecim Gemmis in Veste Aaronis," Epiphanius, 1565. " Gemma- 
rum et Lapidum Historia," Boetius, 1647. Theophrastus " History of Stones 
and Modern History of Gems," Sir John Hill, 1746. "Precious Stones and 
Gems," Streeter. "A Popular Treatise on Gems," Feuchtwanger, 1867. 

3 " A Popular Treatise on Gems," Feuchtwanger, 1867. 

4 "Naturalis Historia," Caius Plinius Secundus, 23 A.D.-J^ A.D, 



532 THE DIAMOND MINES OF SOUTH AFRICA 

commonly octahedral and dodecahedral than cubical, the adamas 
of Pliny may have been the white sapphire crystal, a hexagonal 
prism. 1 

Long before the days of Sindbad the sailor, 2 when the true 
diamond was unquestionably known and prized, and when the 
lucky adventurer filled his pockets with the choicest crystals, 
copper had been substituted for lead in revolving wheels used 
by the most skilful lapidaries for grinding the harder stones ; 
and powdered stone, moistened with oil or water, was sprinkled 
on the grinding wheel or pressed into furrows on its face. Cut- 
ting in the scientific method of the modern art was of compara- 
tively recent development. The grinding or cutting of the 
Indian stones by native lapidaries was, at first, only a surface 
polish of natural planes, and later proficiency did not extend 
beyond an irregular and unsymmetrical fashion, which rarely 
ventured the risk of cleavage. There are perhaps no known 
samples indicating with certainty a higher proficiency in the art 
at the beginning of the Middle Ages than the four large diamonds 
now to be seen on the buckle of the mantle of the Emperor 
Charles the Great, 3 which were planed and polished on their 
natural faces. 4 

There is a particular Oriental cut of diamonds, still followed, 
which had its origin about the year 1000 A.D. This bears the 
distinctive name of " Indian " or " Lustre of India." It had 
four rectangular plates and one upper facet in the form of a 
parallelogram. The stone was polished highly on all surfaces 
except the under side, which was left in its natural state. It is 
thought that the wandering merchants of the East, who travelled 

1 "De Gemmis Plinii," Ernst Friedrich Glocker, 1824. "Precious Stones 
and Gems," Streeter. 

! Ninth century A.D. "Origin of Tales of Voyages," " Cyclopaedia of 
India," Balfour. 

3 " Charles the Great, King of the Franks and Emperor of the Romans," 742 
814 A.D. 

4 " Great Diamonds of the World," Edwin William Streeter, 1882. " Hand- 
book of the Arts of the Middle Ages, ' ' Jules Labarte, London, 1855. 



CUTTING AND POLISHING 



533 



by caravan, brought these stones, or a knowledge of their style, 
from the far Orient to Constantinople, whence they were made 
known to France, Italy, and Holland. 1 

That such forms of gems were made in Paris and in Venice 
as early as the thirteenth century is certain. In 1290 A.D. a 
society of lapidaries was formed at Paris, and at the close of the 
fourteenth century there were professional diamond cutters of 
somewhat higher skill in Nuremburg. In 1365 A.D. an inven- 
tory of the jewels of Luigi d' Angio was made, which mentions 
a diamond having eight facets and another shaped like a shield. 
The facets here spoken of may be only flat sides such as any 
true octahedral crystal presents. 1 ' 

One of the first, if not the first, of European workmen to 
attain any distinction as a diamond cutter was named Hermann, 
living in Paris about 1407 A.D., and it seems to be certain that 
from his time or the beginning of the fifteenth century the busi- 
ness of polishing and developing the diamond became an estab- 
lished industry in western Europe. Gems in the rough were 
somehow finding their way from India and Borneo, and were 
coming into the market not only among kings and the members 
of the royal households but among noblemen and burghers of 
great wealth. In 1465 A.D. there were three registered diamond 
cutters living in the city of Bruges. Perhaps these cutters were 
associated with Louis de Berquem, a native of that city, who an- 
nounced in that year a new method of cutting diamonds and 
established a guild of diamond cutters. 

The method which he pursued and the forms which he 
evolved were deserving the name of a new discovery of which he 
was truly the inventor. With whatever assurance others may 
claim to have invented the art of faceting or of cutting diamonds, 

1 "Precious Stones and Gems," Streeter. "A Treatise on Diamonds and 
Precious Stones," Mawe, 1813. "Treatise on Diamonds and Pearls," David 
Jeffries, 17501751. "On Gems and Precious Stones," Robert Dingley, Phil. 
Trans. Abi. IX, 345, 1747. " Le Grand Lapidaire," Sir John Mandeville, Paris, 
1561. " Les Merveilles des Indes Orientales," etc., Robert de Berguen, 1661. 
" Voyages en Turquie, en Perse et aux Indes," Tavernier, 1676. 2 Ibid. 



534 TH E DIAMOND MINES OF SOUTH AFRICA 

it is very evident that none before him had done so on any sci- 
entific basis of geometrical relations. Berquem was not merely 
a craftsman ; he was an accomplished mathematician, highly 
versed in optical science, and he had determined the true angles 
at which the planes of each facet should lie in reference to its 
crystallization and to its size, in order to make its reflections of 
light most perfect and its color most complete. 

He discovered that in the development of the octahedral 
form there are certain measurements of relation which must be 
preserved in the trimming of the diamond for the perfect re- 
flection of all the light which enters the crystal. By this scientific 
formation he completely changed the basis for estimate of the 
value of diamonds. Under his treatment the diamond of largest 
size and weight was not most valuable, but the gem which was 
transcendent as a light producer or reflector and as a crystal of 
symmetrical parts. The connoisseur, the artist, and the thrifty 
merchant alike have vastly profited by the principles evolved by 
Berquem. He raised the craftsmen of his day from the common 
plane of gem polishers to the higher position of artists and 
skilled lapidaries. The successful lapidary of to-day to whose 
cutting is intrusted the gems of India, Brazil, and Africa must 
be a close student of optics as well as a dexterous stone cutter. 






Fig. i. Fig. 2. Fig. 3. 

Figs, i and 2 above represent the simple octahedral form of 
diamond crystallization. By the second figure it will appear 
that if two pyramids of four triangular sides were joined together 
at their bases, we should have a diamond form with eight trian- 
gular surfaces, or an octahedron. Fig. 3 is the same octahedron 
with its corners either rounded or ground flat as additional 
facets. The diamond's natural edges are not often so straight 
and sharp as here represented, but are usually convex, that is, 



CUTTING AND POLISHING 



535 






bowing outward ; but when mechanically trimmed to perfect their 
shape, each line and angle must be unerringly true. 

Fig. 4 is a cube of six faces having its corners rounded or 
flattened, and Fig. 5 is a double cube or dodecahedron, having 
twelve equal rhombic faces. Some diamonds can 
readily be made to receive these shapes with 
little loss of substance. Fig. 8 represents a 
gem shaped as a parallelogram with a facet 
Fig. 4-' on one upper corner, the lower side showing F 'g-s- 
its natural state. It is called " Indian " or " Lustre of India." 
Figs. 6 and 7 represent the oldest and simplest form of gem 
cutting, called the " table cut." It suits the other precious gems 
much better than the pure dia- 
mond. A celebrated " table " a/_ 
diamond was given by Prince 
George (afterwards George IV.) 

Fitzherbert. She had it split along the line from 
a to b, and used each half to fit in the face of a locket ; one 
holding her own portrait, and the other 
that of her princely lover. The diamond 
with the portrait of Mrs. Fitzherbert was 
buried with the old king in the locket 
which hung on his neck. 1 

The first and simplest modification of 
the " table cut " of a diamond is called the " Old English sin- 
gle " or the "star single cut." By this arrangement the table 
cut diamond had its top part planed down about the edges to 
represent an eight-pointed star whose centre figure was an 
octagon, or elongated octagon, if the stone was longer than its 
width. This style of cutting appears in sets of old diamonds 
for crown jewels or ordinary wear. These sloping triangular 
faces were ground upon the edge of the upper surface of the 
stone only, reaching from the flat part, which is then technically 
called the "table," to the central line which is called the 
"girdle," and these cut surfaces are called "facets" or small 
1 "Macaulay's Essays," Thomas Babington Macaulay. 



Fig. 8. 



536 THE DIAMOND MINES OF SOUTH AFRICA 



faces. Their size and shape are most accurately measured and 
most exactly ground. 

Figs. 9 a, 9 b, 9 c represent, successively, side or girdle, top 
or table, and back or culet of the next most simple cut of 






Fig. 9 a. l''^. 9 /'. Fig. 9 c. 

modern date, which is of a scientific cast. It is called the 
"single cut brilliant," a modification of the simple table cut. 
Fig. 10 below represents a single cut having sixteen triangular 
facets on its upper section and twelve facets on the under 




Fig. 10. 





I-'i". 12. 



section, plus eight long facets. Figs. 11 and 12 show one of 
half that number, but both belong to the style here described 
the single cut brilliant. Indeed, with very small stones, the 
single cut has but four faces above and four below. In com- 
mercial circles they are called " single sets." 





Fig. 13. Fig. 14. 

The two figures above present another modification of the 
simple table cut of India. It is called the "step cut." In this 
style the plane above the girdle is only half as thick as that 
below the girdle. From Fig. 13 it appears that the part 
above the girdle has been bevelled off at two different angles, 
making two "steps" besides the table. The other figure 



CUTTING AND POLISHING 537 

represents the part beneath the girdle, which being twice as 
thick as the upper section is cut with six steps instead of 
three. In closely studying this step cut and the table cut it 
was discovered that the diamond crystallized in thin laminae 
or plates, and that it might be split into very thin sections 
resembling plates of mica. By taking advantage of these 
" lines of cleavage," as they are called, many large diamonds 
were split into thin leaves and used as faces of small pictures 
enclosed in lockets. At Queen Victoria's coronation this thin 
sheet diamond was so common that many distinguished guests 
were favored with a gift of their own likene.sses encased in golden 
frames and covered with a diamond instead of a glass face. 

In the plates below, Fig. 15 and Fig. 16, facets of the "rose 
cut" pattern are represented. It will be seen that the bottom 
of the diamond is flat, though not unpolished, 
while all the facets lie above the girdle. This 
design, which is called the " Hol- 
land," groups twenty-four facets, 
but a simpler style known as the 
Fig. 15. "Antwerp rose" shows facets Fig. 16. 

ranging from six to sixteen. This rose cut is a very convenient 
style to adopt for fragments which have been cleft from large 
stones, or for diamonds which are imperfect in their crystal- 
lization on one side. If well proportioned, the depth of the 
rose must be one-half its breadth at the base. 

In the rose cut diamond every facet is a triangle and all 
meet at the central apex, forming a cupola. When the facets 
on large stones number thirty-two, the dealers call it " fiam 
minghi " or " half brilliant." A common practice of the trade 
is to obtain a second " fiam minghi " of the same size, but cut 
in quartz crystal or even in glass, and glue their bases together 
with gum mastic, thus forming the " briolet " or " brilliolet," 
which is palmed off for a pure diamond. Briolets are pear- 
shaped or oval stones, having neither table, culet, nor edge, but 
covered with triangular-shaped facets, sometimes pierced at their 
points of greatest diameter, to be suspended on an axis. 





538 THE DIAMOND MINES OF SOUTH AFRICA 



It has been told how the diamond by Berquem's talent was 
first cut in harmonious and systematic proportion and regular 
facets at such an angle to its axis and to each other that the 
fullest play of reflected light is secured from every surface on 
which it strikes. His art produced the single cut brilliant, the 
highest achievement of the lapidary of his day. Near the close 
of the seventeenth century a Venetian engraver, named Vincenzo 
Peruzzi, while experimenting to get rid of obnoxious color in 
small diamonds, invented the double faceting which is now known 
as the "brilliant." It is regarded as the perfection of the lapi- 
dary's art, and is adopted in cut- 
ting the most costly gems now 
put upon the market. There are 
thirty-two facets in its upper sec- 
tion, and twenty-four below the 
girdle. A diamond cut in this 
style is shown in Figs. 17 and 18. 
The usual double cut brilliant has only fifty-six facets, but, 
of late years a supposed improvement has been made by adding 
eight star facets around the culet, which makes a total of sixty- 
four facets. The proportions of measurements for the perfect 
brilliant diamond do not hold for other colored gems whose 
depth increases or diminishes their color. The triangular facets 
on the bezel, which touch the table, are named " star facets," 
while those which touch the girdle are " skill facets.'" 




Fig. 17. 







Fig. 19. 





Fig. 21. 



In order to show the names of lines and the geometrical re- 
lations of diamonds as a lapidary sees them, the above figures 
may prove helpful. Fig. 19 shows the side view of an ordi- 
nary octahedral or eight-sided diamond. Fig. 20 shows first the 



CUTTING AND POLISHING 539 

line at a cutting off the upper point of the diamond. When 
this is accomplished, the flat top surface is called the " table." 
The line at c, which is the largest girth of the diamond, is called 
its "girdle." The space b, between the girdle and its "table," 
is called the " bezel." The line at e cuts off the sharp lower 
point, and its flat surface is the " culet." The space between 
the culet and the girdle is called the " pavilion." 



Cleaving Diamonds 

There are three distinct processes in the treatment of dia- 
monds by the lapidary cleaving, cutting, and polishing. To 
split the diamond successfully demands a thorough knowledge 
of its individual character as well as of its generic crystallization 
and lines of cleavage. The skilled lapidary takes in hand a large 
rough diamond. If it is an Indian or Brazilian stone, it is coated 
or partly coated with a hard dull crust. Its corners are perhaps 
abraded. It may have defects or cracks in its surface, unequal 
coloring, or black deposits in its interior. He must needs re- 
move the crust, correct the distortion of the crystal, remove or 
conceal its defects, and decide what is the largest perfect gem 
which can be cut from the rough stone. He must be able to see 
the priceless jewel through its shrouding veil, and determine on 
which surfaces of the stone its prominent corners must rest. 
Having decided what shape will best befit the stone, he must 
know whether the rejected portions can be split off safely or 
whether they must be ground off. Grinding away the rejected 
portions is probably the safest procedure, but it is the slowest 
and most expensive. The quickest method is to split off the 
surplus material. The process will be easy if the proposed frac- 
ture is in the direct line of cleavage in that particular stone. If 
not, his attempt at splitting may ruin a gem of countless price. 
Shall he make the attempt? He must be both an expert and a 
man of nerve. If he be so, a single feat of successful polishing 
may bring him fortune and the reputation of a master, while a 
single disastrous venture may quite undo him. 



540 THE DIAMOND MINES OF SOUTH AFRICA 

The early lapidaries dared not attempt the splitting of a stone 
to correct its faults or alter its natural form. Every stone was 
estimated according to the impression it made upon the scales. 
Hence its facets were only smooth flat surfaces ground upon the 
rounded exterior, an unmitigated rose cut of trivial triangles, 
or a terraced surface of rings and bands. The master of his 
craft to-day must make his diamonds perfect reflectors of light at 




Cleaving. 

all hazards. If any excrescence exists,he must cut it away, or the 
light which enters a flattened surface may be so entangled that it 
will never emerge. When he takes up a cross-grained, defective 
stone,he will reject it. Like a true surgeon he will quickly dis- 
cern how he may remove most safely a defective part, and will 
proceed boldly with his task. His first step in the work is to 
scratch the surface round the part to be split off with another 
diamond. Having made the diamond fast in a cement bed com- 



CUTTING AND POLISHING 

posed of brickdust and resin, he applies the edge of a steel knife 
to the scratched surface, and strikes a quick, hard blow with a 
slender rod. If he has struck the lines of cleavage, the external 
scale is at once removed, for the diamond, despite its hardness, is 
quite easily fractured. Then the split surface must be polished. 
If no other scale or marked inequality needs removing by 
splitting, the next operation is that of grinding. 

Cutting 

As the diamond is the hardest of all known substances, it is 
evident that much patience and strength are required in reducing 




Cutting. 

its size or altering its rough figure by grinding. The ordinary 
file would serve to reduce some other gems, but it will not touch 
the diamond. Diamond cut diamond is not merely a current 
phrase, for diamond dust is now invariably used in polishing or 
grinding this precious stone. 

For cutting and polishing purposes the lapidary has a table 
above which a flat steel wheel revolves horizontally. On the 
upper surface of this wheel are fine grooves or striae, cut angling 



542 THE DIAMOND MINES OF SOUTH AFRICA 

from its centre to its perimeter. By means of belts beneath the 
table, the grinding wheel is made to turn at a rate from three to 
four thousand revolutions a minute. Diamond dust mixed with 
olive oil is applied to the upper face of the wheel, and against 
this erosive surface is held the diamond to be ground or cut. 

For this object the diamond is set in a fusible solder on the 
end of a copper cupel which is held firmly against the surface of 
the wheel by a small projecting arm and clamp. By adjusting 
this holder, the lapidary presses the exposed face of the stone on 
the revolving wheel until the desired amount of material has 
been ground away and the proper angles turned. Such work in 
its finishing stages cannot be intrusted to a tyro or experimenter. 
Unusual patience and steadiness of nerve are required for such a 
task. When the facet is finished, the workman wipes the dust 
off and tests its smoothness and finish, after which he resets the 
diamond, leaving the uncut facet exposed which he intends to 
cut next. 

Most of the " skill " facets and " underskill " facets are made 
by grinding, while the lozenge and larger faces are first shaped, 
when possible, by cleaving. If the stone is thick enough to 
form a brilliant, the lapidary first forms the table, and then suc- 
cessively the adjacent facets and lozenges. The table must be 
absolutely flawless and smooth, while all the surrounding facets 
in an ideal brilliant must hold the same precise angles and have 
their shape correspond to the thousandth of an inch. After 
completing the bezel, the pavilion is next developed. The 
underskill facets of the pavilion must match exactly at the girdle 
with those of the bezel, and the girdle when finished should be 
as sharp as a knife. Some lapidaries leave the girdle blunt, but 
with a great sacrifice of brilliancy in the gem. The triangles 
and lozenges of the pavilion must, of course, be much larger 
than those of the bezel. 

There is a still simpler method of cutting diamonds by a 
device attributed to Berquem. Two uncut stones are cemented 
into the ends of two sticks resembling penholders in shape. 
Then the operator grasps these handles and presses the stones 



CUTTING AND POLISHING 



543 




544 THE DIAMOND MINES OF SOUTH AFRICA 

against each other with a rubbing motion over a trough. Con- 
siderable leverage is obtained for the rubbing by resting the 
holders against projectors at the sides of the trough. The ex- 
posed face of the stones is coated with diamond dust to advance 
the process. In this laborious way facets may be ground, and 
the cutting may be completed by repeatedly refixing the stones 
in the cement. Expert handling is necessary to keep the dia- 
monds from becoming overheated by the constant friction. 



Polishing 

The third process is that of polishing. The method em- 
ployed does not differ materially from that adopted in cutting, 
described above ; but as this is the finishing process, all irregu- 
larities in faceting must be corrected and the practised eye of the 
artist must detect and remedy every defect. Each line and angle 
must be made geometrically correct; each facet and lozenge must 
be shaped to perfection. The colorless stone must glisten pure 
as a dewdrop sparkling in the sun, producing the colors of the 
prismatic spectrum ; the gem of red or blue or green color must 
flash forth its hue with intense brilliancy. 

Such exact and delicate alignments are not the work of a 
day, though the time required has been greatly shortened by 
modern methods. The patience of weeks and even of months 
must be expended in perfecting these tiny crystals. It is said 
that it was the work of two years to cut the celebrated Pitt dia- 
mond, now among the French jewels, and the lapidary received 
for his skill and labor the sum of ^3500 or 117,500. The last 
cutting of the Koh-i-nur by Coster of Amsterdam in thirty-eight 
days was unusually rapid. The ablest workmen in Holland 
were engaged continuously on it and the wheel was driven by 
steam power ; yet it cost 140,000 to do the work and the 
diamond lost eighty-four carats in weight. 1 

1 " Great Diamonds of the World," Streeter, 1882. 



CUTTING AND POLISHING 



545 




546 THE DIAMOND MINES OF SOUTH AFRICA 

Loss in Cutting 

No general rule can be stated covering loss which occurs in 
cutting gems. The waste depends on the character of the stone, 
its perfect natural form and crystallization, its purity, and 
the style of the cut adopted. Perfect octahedrons lose two-fifths 
of their weight, if cut as brilliants. Rhombohedrons will lose 
over half of their weight in taking the same form, and stones of 
other shape will lose as much or more. The following figures 
will show at what cost of substance some of the natural gems 
have been perfected in the process of their cutting: The 
Mogul in its rough outer coat weighed originally 780^ 
carats ; when cut it weighed only 279^ carats, a loss of nearly 
two-thirds. The Regent weighed 410 carats and was reduced 
to 136^ carats. The weight of the Koh-i-nur was originally 
793 carats. It was first cut unskilfully by Hortensio Borgio to 
1 86^ carats, and a second cutting reduced it to 102^- carats, a 
loss by both processes of the astonishing amount of 690^- carats, 
or more than six and a half times its present weight. The 
L'Etoile du Sud shrunk from 254^ carats to 124^% carats in 
the process of cutting. 1 The average loss of South African 
diamonds by cutting is from one-half to three-fifths of their 
gross weight. The 428^ carat diamond found in De Beers 
mine lost 200 carats in cutting. 

It has been demonstrated in cutting that diamonds are of 
different degrees of hardness and that the same stone may 
exhibit different degrees on different faces. The Koh-i-nur is 
a signal example of this fact. In cutting the facet near a yellow 
flaw, the section grew noticeably hard, until six hours' grinding 
at a speed of 2400 revolutions a minute produced only the 
faintest change. A speed of 3000 revolutions was necessary to 
cause any perceptible loss of material on that facet. A speed of 
4000 revolutions a minute is about the average now in vogue 
at Amsterdam. 2 

1 " Famous Diamonds of the World," Streeter, 1882. 

2 Description by Messrs. Veder and Rozelaar, dated 6th March, 1902. 



CUTTING AND POLISHING 547 

There is another material loss occurring in cutting or in the 
handling of rough diamonds from a curious infirmity of some of 
these crystals. The explosion of diamonds sometimes occurs, 
and the loss is the greater because large stones are more liable to 
explode or fly into pieces than small ones. This phenomenon 
is attributed to the heat of the hot solder, or frictional heat of 
the revolving disk. 

The Lapidaries 

Early handlers of the diamond were hardly more than pol- 
ishers, striving to produce an even, glistening surface, and satisfied 
to retain the natural face of the stone, or to grind away some 
upper portion of the crust. This clearly appears from the many 
old, half-polished stones that have been found in treasuries of 
gems. A signal instance is shown on the royal mantle of Charle- 
magne, still preserved in the French National Collection. In 
the clasp of this robe are diamonds whose natural octahedral 
faces have been simply polished. In ancient church furnishings 
diamonds have been found with an upper table and four polished 
borders, and the lower sides cut as four-sided prisms or pyramids. 
Streeter quotes this inventory of the Duke of Anjou's jewels 
exhibited in 1360 A.D. : (i) a diamond of a shield shape, from a 
reliquary ; (2) two small diamonds from the same reliquary, with 
three flat-cut, four-cornered facets on both sides; (3) a small 
diamond in the form of a round mirror ; (4) a thick diamond 
with four facets ; (5) a diamond in the form of a lozenge ; (6) an 
eight-sided, and (7) a six-sided plain diamond. 1 We must allow, 
of course, for the mistakes and the ignorance of those who may 
have catalogued rock crystals for diamonds, but granting that 
some were diamonds, their existence shows what forms were then 
prevalent and the real development of diamond cutting. 

Previous to the success of de Berquem as a lapidary, there 

were polishers and cutters in Paris and at Nuremburg, as has 

been noted. A guild was organized in Paris in 1290 A.D., and 

the table cutters joined in a guild with the stone engravers in 

' "Precious Stones and Gems," Streeter, 1880. 



548 THE DIAMOND MINES OF SOUTH AFRICA 

Nuremburg, which became a primitive Lapidaries' Union. They 
received and taught the apprentices on the strict condition of 
their contract to serve for five or six years before undertaking 
business for themselves. The artist Hermann won for himself 
an honorable name in France as early as 1407 A.D., and Guten- 
berg, the originator of the art of printing from block type, learned 
the lapidary's art at Strasburg in 1434. At a dinner given by 
the extravagant Charles the Bold, the Duke of Burgundy, to the 
king of France, he bestowed upon his guests eleven diamonds 
partially cut and mounted in gold. Undoubtedly these jewels 
were declared in the height of the fashion by European artists. 
But it is from de Berquem's day that the profession reckons its 
firm establishment, and his contemporaries have acknowledged 
him as the father of his art. 

Though many Europeans have become skilful workers at 
this trade, the most successful lapidaries have been of Hebrew 
stock. The Jews had, at one time, the monopoly of the trade 
in diamonds in Portugal, and their especial centre of business was 
Lisbon. The old " Lisbon cut " of diamonds has never been 
surpassed for perfection and beauty of workmanship. But un- 
fortunately for Portugal and for the Jews, religious bigotry 
kindled the fires of persecution against this ancient people, and 
they were expelled from the kingdom. Hospitable little Hol- 
land opened her doors to receive the exiled merchants and 
lapidaries, and Amsterdam has since become the central mart 
for the diamond merchant and his comrade, the diamond cutter. 
Out of thirty-five thousand Jews who reside there, at least a 
third are engaged in one department or another of the diamond 
industry. 

The settlement of some of de Berquem's pupils at Amster- 
dam was probably the reason why the exiled Jews selected that 
city as their home; but others went to Antwerp and some made 
their homes in Paris. The descendants of these expatriated 
Jews received especial encouragement and protection from Car- 
dinal Mazarin two hundred years later. Twelve of the largest 
crown jewels were intrusted to them to be recut on Berquem's 



CUTTING AND POLISHING 



549 




550 THE DIAMOND MINES OF SOUTH AFRICA 

principle. Their success was so marked that these stones were 
afterwards known as the " Twelve Mazarins." Unfortunately 
these rare gems were poorly guarded and all but the tenth had 
disappeared by 1791. The French cutter Jarlet gained an inter- 
national reputation in the seventeenth century by cutting one of 
the notable jewels of the Russian crown weighing 90 carats, but 
the industry withered in France in spite of its special encourage- 
ment by Mazarin and other powerful ministers. 

England and Holland had secured almost exclusive trade 
relations with the East, from whence the diamond supply was 
obtained. Hence the Hebrews of these countries secured con- 
trol of the diamond industry, and French lapidaries sought 
employment in vain. Then the revocation of the Edict of 
Nantes flooded Holland with French refugees. Of the seventy- 
five diamond cutters whom Mazarin had so carefully guarded, 
only five remained in 1775. Inquiry showed that the total rough 
diamond stock in Paris, just before the outbreak of the Revolu- 
tion of 1789, was only 3832 carats, and there was little employ- 
ment obtainable in recutting old stones. During the Revolution 
and the troubled Napoleonic reign, the industry was fatally 
paralyzed, and diamonds were sent out of France to Antwerp for 
cutting. 

In the eighteenth century there was a determined push in 
England to foster the diamond-cutting industry, and some ex- 
pert workmen, headed by Ralph Potter, made a stout struggle 
to hold the home trade. The so-called "Old English style" 
was developed on strict mathematical lines, and gems cut by 
these artists are still eagerly sought as models of the lapidary's 
art; but the centralizing drift to Holland was too strong for 
competition until the discovery of the South African Diamond 
Fields. In the last twenty years the languishing art has 
raised its head in England, and become, without doubt, a well- 
established industry. A hundred and fifty years ago London 
was accounted the chief centre of business for lapidaries, and 
it is not beyond expectation that its former preeminence may 
be reestablished. Even now it is thought that diamond crystals 



CUTTING AND POLISHING 551 

are cut in London town as well as the work can be done in Am- 
sterdam. 

In the United States the late Henry G. Morse of Boston 
was the pioneer in establishing the lapidary business on a suc- 
cessful footing early in the last (nineteenth) century. He opened 
his workshop in Boston in 1866, and made several important 
improvements upon the cumbrous machinery in use in Europe. 
His business was confined, at first, to recutting and polishing 
damaged gems; but the influx of South African diamonds brought 
about speedily an enlargement of his works and the employment 
of thirty expert hands. At the start only foreign workmen were 
engaged, but Mr. Morse succeeded in training American women 
to a height of proficiency as lapidaries which rivalled the best 
foreign work. Among the fine gems cut and polished in his 
shop were four weighing fifty carats each, and he later scored a 
notable success with the cutting of a superb South African dia- 
mond weighing 125 carats. The brilliant fashioned from this 
stone weighed 77 carats, and has been greatly admired by con- 
noisseurs as a specimen of exquisite beauty and purity developed 
by perfect workmanship. The cutting and finishing of this gem 
was a work occupying three and a half months. 

In spite of this well-designed and ably pushed venture of 
Mr. Morse, American lapidaries have struggled continuously 
under serious handicaps. The United States is not a producer 
of diamonds, and Europe is the established mart for rough stones 
from India, Africa, and Brazil. Moreover, the business of dia- 
mond cutting has been so firmly rooted in Europe that the work 
naturally gravitates to these older establishments. Foreign lapi- 
daries and dealers enjoy a further advantage in the fact that the 
banks of England and Holland make loans on uncut stones, 
knowing that the finished diamond is much enhanced in value, 
while American bankers do not grant such assistance to Ameri- 
can cutters and dealers. Nevertheless, the work of diamond 
cutting has been so persistently developed that over half of the 
diamonds imported now enter as rough stones. 1 

1 George F. Kunz. 




A Group of Directors, General Manager, and Secretary of De lieers Consolidated Mines Ltd. 




CHAPTER XIX 

AN UPLIFTING POWER 

HAT a change came over the dismal face of 
South Africa with the discovery and develop- 
ment of the diamond mines ! In a former 
chapter it has been shown how dragging had 
been the advance from the few scattered settle- 
ments on the coast up to the year when the 
revelation of diamonds drew the first rush of prospectors to the 
banks of the Vaal. The yield of this marvellous field was, 
from the first, of material consequence in the sum of South 
African products, but it was of far greater importance in the 
stimulus which it gave to the flagging and stinted enterprises 
and the sinking hearts of the colonists. 

The bits of iron hoop that scraped the diamond-bearing 
ground were as transforming as magicians' wands. The river 

552 



AN UPLIFTING POWER 



553 



wash of the Vaal glittered like the diamond-strewn valley of 
Sindbad. No Man's Land had the sparkle of diamond founts. 
No part of the world was too remote to be dazzled by the vision 
of the novel Golconda, and the black face of the despised karoo 
changed in a twinkling to one of transcendent promise. 

Then came the rush from every quarter of the globe of 
ardent visionaries and fortune-hunters, streaming over the desert 




Natives riding Bullocks. 

sands of South Africa from every coast port to the Diamond 
Fields, while from far inland the tribesmen flocked to the same 
glittering beacon. Bitter experience rubbed the glamour from 
the eyes of thousands of visionaries, who trooped back dis- 
heartened, but the plucky and the lucky held their ground and 
thousands came streaming in to take the places of the faint- 
hearted. It followed naturally, too, that thousands who would 
never have come to South Africa except at the beckoning of the 



554 THE DIAMOND MINES OF SOUTH AFRICA 

diamond lure, remained in the country even after the blasting 
of their hopes of diamond winning. Many were ashamed to 




Pioneers Trekking. 

run away with the confession of failure ; many were too poor 
to get away, and many were keen to see the profitable openings 
in other occupations for their work and savings. So every 
industry in the colonies gained new headway with the influx 




A Prospecting Expedition to Mashonaland. 

of capital and labor. Supplies of all kinds were needed for the 
bustling diamond camps and the flow of travel between the 
mines and the coast. 



AN UPLIFTING POWER 



555 




Trekking on the Veld. 

This demand was quite enough to stir the pulse of produc- 
tion in every part of South Africa, and the heartening impulse 
thus given was sustained and advanced far beyond the stretch 
of this novel requirement by the rising faith in the possibilities 
and future of South Africa as a field for investment, which now 
began to lift the drooping spirits of the colonists and to attract 
the cooperation of the home country and the leading nations 
of the world. Hopeful prospectors rambled off farther and 
farther over the deserts and ranges, or followed the water- 
courses, testing the sands for diamonds or gold, and picking 
at every promising ledge in their search for ore. Pioneer 




Trekking on the Veld. 



556 THE DIAMOND MINES OF SOUTH AFRICA 

graziers trekked to new pasture grounds with their flocks and 
herds. Abandoned farms were reoccupied and virgin soil 
upturned for crops. Manufactures of various kinds began to 
spring up and multiply. Not only Cape Town but little coast 
ports were thick set with steamers busily discharging cargoes 
on piers or in lighters and bidding for exports at rates highly 
stimulating to the products of the Colonies. 

The march of development was signally marked in the con- 




The first and only Load of Cotton raised in South Africa. 

struction of railways to meet the pressing demands for inland 
communication and transportation, and especially the imperative 
call of the Diamond Fields. The progress of mining was 
greatly handicapped from the start by the heavy cost of drag- 
ging supplies in lumbering ox-wagons for hundreds of miles 
from the coast ports, and the patent impossibility of moving 
any large plant in this way for mine opening or diamond win- 
ning. The pioneer railway from Cape Town to Wellington 
barely covered a twelfth of the stretch from the coast to the 



AN UPLIFTING POWER 



557 



mines, and the little lines from Salt River to Wynberg, and 
from Port Elizabeth to Uitenhage were of no service in the 
advance of transportation. 

However, capital was wary and loath to invest in any of the 
projects for railway building into the heart of South Africa, until 
the continued working of the Diamond Fields for three years 
convinced investors that rich diamond deposits were indeed 




Zebra. 



open, whose continuance in depth might reasonably be antici- 
pated. When the export of diamonds in 1872 amounted to 
over ; i, 000,000, the Cape Government authorized the pur- 
chase of the sixty miles of railway then in place in the colony 
and sanctioned the extension of the existing line and the con- 
struction of railways from Port Elizabeth and East London. 
It was a heavy strain to raise capital for the extension of all 
simultaneously ; so the advances were groping and slow. At 



558 THE DIAMOND MINES OF SOUTH AFRICA 




Coach leaving Kimberley for the Transvaal Gold Fields before the Day of the Railway. 

some points, too, there was even an obstinate fight against any 
railway extension in their direction. Little towns that had been 
centres of distribution for surrounding districts feared that the 




Cape Town and Table Mountain from Table Bay. 



AN UPLIFTING POWER 



559 



railroad would divert some trade and shake their preeminence. 
It was further contended that the whole scheme was chimerical 
and that the uncertain yield of the South African farms and the 
wide-ranging pasture lands could not pay interest on the cost 
of construction and maintenance. One of the most persistent 
objectors, the town of Worcester, actually sent in five petitions 
against the extension of the railway line to that point. 




A Group of Well-known Kimberley Men. 

There was further protracted disputing over the proper gauge 
for adoption after an extension had been determined. The short 
line already constructed from Cape Town to Wellington was of 
the standard English gauge, 4 feet 8J inches, but the continuance 
of this gauge was generally opposed. It was urged that the light 
traffic of the country would not warrant the heavy outlay requisite 
for the construction of lines of the standard adapted to the require- 
ments of a thickly settled country like Great Britain, and that only 
narrow-gauge lines were practicable. Some would have pushed 
this reduction of gauge to 2 feet 6 inches, or even less, in view 
of the fact that some of the lines might be reduced to the necessity 
of resorting to tram car and mule service, but the final conclusion 



560 THE DIAMOND MINES OF SOUTH AFRICA 

was the adoption of a standard gauge of 3 feet 6 inches. It was 
properly recognized that uniformity of gauge, at any rate, was 
essential for intercommunication, and whimsical notions of con- 
struction were not suffered to break this uniformity. Time has 
shown the fallacy of these pessimistic predictions as well as the 
adoption of the 3 feet 6 inches gauge. 

There was, however, one essential error in the whole scheme 
of construction. The pressure of the demand of widely separated 
points for railway construction was so hard to resist that the Par- 




Eland. 



liamentary authorization for railway extensions was far in excess 
of what was feasible at the time in view of the limited capital that 
could be secured for the prosecution of the scheme. The rivalry 
of the principal ports was too keen to permit of the drafting of any 
cooperative plan of extension, for the superior accommodation, 
even temporarily, accorded to any one port would be challenged 
by others as injurious favoritism. So, instead of carrying forward 
a single main line by the most direct or feasible route to the Dia- 
mond Fields to meet the most pressing demands for communi- 
cation, there was for many years only a crawling advance from the 



AN UPLIFTING POWER 



561 




562 THE DIAMOND MINES OF SOUTH AFRICA 




Table Mountain, from the West. 



competing coast ports, Cape Town and Port Elizabeth. Neither 
of these competing lines were able to pay any adequate return 
upon the capital invested, and the common aim of reaching the 
Diamond Fields was blocked and greatly delayed. Kimberley was 
distant only 485 miles from the nearest outlet on the coast, Port 
Elizabeth ; but 1600 miles of converging railway lines were actu- 
ally built before one was extended from De Aar to Kimberley, 
in November, 1885, then first putting the richly productive 
diamond mines in railway communication with the coast. 

All the lines in operation at this time were single lines, with 
the exception of the Cape Town-Wynberg line, and the first six 
miles of the Port Elizabeth-Uitenhage line. The most difficult 
engineering in the course of this railway extension was in the 
crossing of the barrier range of mountains forming the ridge of 
the karoo plateau. After repeated surveys an entrance for the 



AN UPLIFTING POWER 



563 



line from Cape Town was effected through the Hex River Valley 

with a gradual ascent to Hex River Kast, where the line begins 

to climb the mountains by sweeping curves and zigzags, piercing 

some of the spurs in tunnels, and spanning gulleys with viaducts, 

until it attains its highest elevation of 3588 feet at Pieter Meintjes 

Fontein, 77 miles from Worcester. This is a trifle higher than 

the summit of 

Table Mountain, 

which rises in air 

3582 feet above 

Cape Town. For 

a stretch of more 

than 20 miles in 

the ascent of this 

ridge, the gradients 

are one in 40 and 

one in 45, with 

curves of five chains 

radius. 

In the year fol- 
lowing the exten- 
sion to Kimberley 
there was a fortu- 
nate impulse to the 
extension and oper- 
ation of all the lines 
by the discovery of 
the Witwatersrand 
Gold Fields. Then first appeared some substantial prospect of 
profit for all the competing lines by the addition of another great 
centre of attraction and production. The junction of the Cape 
Town and Port Elizabeth line at De Aar, in March, 1884, had 
largely diverted the flow of freight and passenger traffic between 
the Diamond Fields and the coast, which, for some years, had 
been passing principally along the line through Graaff Reinet; but 
the rise of Johannesburg offset this loss to the Port Elizabeth, 




His Excellency Lord Milner, Governor of the Cape Colony. 



564 THE DIAMOND MINES OF SOUTH AFRICA 

Graaff Reinet, and East London lines. The linking of the Dia- 
mond and Gold Fields by direct railway communication from 
Kimberley to Johannesburg was apparently of high importance ; 
but this extension has been blocked for years by the action of the 
Orange Free State in refusing to build the line themselves or to 
allow either the Cape Government or private corporations to 
construct it. Several years were passed in dilly-dallying before 
sanction was given to the Cape Government for an extension of 
a connecting link of the Cape Town and Port Elizabeth lines 




Donkey Transport. 

from Naauw Port across the Orange River at Norvals Pont, 
thence to Bloemfontein and the Vaal River at Vereeniging, where 
it connected with the Netherland Company's system of the 
Transvaal. It was not until September, 1892, that the first 
through train from the Cape reached Pretoria, but after the 
essential link was constructed, the Cape Town and Port Eliza- 
beth lines contrived to secure the greater share of the Transvaal 
traffic for the next three or four years. 

While these two lines were delayed in reaching out for the 
business of the Gold Fields, a more favored competitor, the 
Netherlands Railway Company, was actively building an eastern 



AN UPLIFTING POWER 



565 



line from Portuguese territory through Middleburg to Pretoria, 
and shortly afterward running radial lines southeast to Natal 
and southwest to Klerksdorp. By the extension of these well- 
designed lines the first through train from Delagoa Bay arrived 
in Johannesburg in November, 1894, a "d the first train from 
Natal in December, 1895. More than two years later, in 
March, 1898, the dragging extension of the Graaff Reinet line 
was opened to Rosmead Junction, on the main line to Port 
Elizabeth, and was then in a position to assist in carrying 
merchandise from the coast to the Diamond and Gold Fields. 




A South African Farm. 



While these railway extensions essential to the development 
of the existing States and Colonies in South Africa were more 
or less efficiently accomplished, the grand project of Mr. 
Rhodes for a railway running far north into the heart of Africa 
was most energetically prosecuted. By the advance of his 
exploration and colonization plan, to be hereafter described, the 
range of British territory was extended from Table Bay to the 
shores of Lakes Tanganyika and Moero. 

The line from Kimberley was opened to Vryburg, 774 miles 
from Cape Town, December i, I89O. 1 Thus far the conserva- 
tive government was prevailed upon to proceed, but the profit 
1 Report of the General Manager of Railways, Cape of Good Hope, 1898. 



566 THE DIAMOND MINES OE SOUTH AFRICA 



from any further extension seemed so essentially speculative 
that it is very doubtful if any further advance would have been 
made, had it not been for the daring enterprise of the Bechuana- 
land Railway Company, an organization promoted and financed 
by Mr. Rhodes and his far-sighted associates. Following hard 

upon the heels of 
the pioneers in 
Mashonaland and 
the conquest of 
Matabel eland the 
line from Vryburg 
was opened to 
Bulawayo in No- 
vember, 1897. 

When the 
grand importance 
of this railway 
advance became 
clear, even to the 
doubters, the Brit- 
ish Government subsequently guaranteed a loan of ^2,000,000 
to carry the line 800 miles farther on to Lake Tanganyika. 

With the rate of progress attained it was expected that Aber- 
corn at the foot of Lake Tanganyika would be reached in four 
years, but the outbreak of the war with the South African States 
was an unlooked-for clog to this advance. As soon as the line 
has reached Lake Tanganyika a further extension of 600 miles 
to Uganda through the Congo Free State has been guaranteed 
by an appropriation of the needed funds by vote of the share- 
holders of the African Transcontinental Railway Company. 
Besides this main line of advance, the Beira Railway, which 
was constructed with a gauge of two feet, had been completed 
and engines were running as far as Salisbury over a stretch of 
line 375 miles in length before the close of 1900. The narrow 
gauge of two feet was soon found to be unworkable, and the line 
has already been relaid from Beira to Umtali with heavier rails 




Natives shearing Sheep. 



AN UPLIFTING POWER 



567 



and with the standard South African gauge of 3 feet 6 inches, 
the remaining stretch from Umtali to Salisbury having been 
originally laid with the broader gauge. 

In spite of the lack of cooperation and capital, and all other 
impediments and delays in view of the character of the country, 
the advance of railway systems in South Africa has been phe- 
nomenal in the last few years. Including the six or seven short 
private lines constructed for the advance of mining operations 
and suburban and other local traffic, there were 2264 miles of 
railway in Cape Colony at the close of the year 1898. The 
Transvaal came next with 777 miles, followed by Rhodesia with 
604 miles, Natal with 465 miles, and the Free State with 361 
miles. Besides this aggregate, 256 miles had been constructed 




Dutch Farm in the Karoo. 



in Portuguese territory, making a total of 4727 miles of railway 
actually opened and working in South Africa, and more than 
half as much more in process of construction, or guaranteed by 
appropriations. 

In the struggle to reach the goal of the Diamond Fields, with 
the handicap of the lack of capital, it is not surprising that much 
of the roadways and the rolling stock fell below any high modern 
standard. The light rails and rickety cars answered the purpose 



568 THE DIAMOND MINES OF SOUTH AFRICA 




Beaufort West, a Little Town on the High Veld. 

of the day, however, fairly well, and have since been largely 
replaced by a plant that will bear wear and tear, but is still not 
up to the requirements. Makeshift bridges were soon sup- 
planted by durable structures, and other engineering works on 
the lines of the various systems have also been greatly improved. 
The engineers who advanced the pioneer lines deserve, on the 
whole, high credit for their energy and talent in piercing or trav- 
ersing the barriers in their way. At the time when the first 
train reached the Witwatersrand Gold Fields, at the close of 
1892, there were somewhat more than 8500 bridges, culverts, and 
cuts to be counted on the various lines. Some of this bridge 
construction, especially the bridges across the Orange and Vaal 
rivers, was of a high order of excellence. The Orange River 
bridge on the Kimberley line has a length of 1230 feet, with 
open spans of 130 feet each between the piers. The Bethulie 
bridge is 1486 feet long, and the Norvals Pont bridge, 1 the 
longest of all, has 13 spans of 130 feet, and a total stretch 
closely approaching 1700 feet. The total cost of this fine 
bridge was ^76,593. At Fourteen Streams, 2 on the Vaal River, 
there is a bridge often spans of 133 feet that is fittingly classed 
with the chief Orange River bridges. 
1 See photograph, Chapter XX, of this bridge after destruction by the Boers. 2 Ibid, 



AN UPLIFTING POWER 



569 



For rapidity of building railways the palm has heretofore been 
claimed by America, but the best American records have been 
challenged repeatedly in the advance of the African Transcon- 
tinental Railway, and it is now claimed that the world's record 
for rapid construction and bridge building has been captured by 
The Patent Shaft and Axletree Company, of Wednesbury, Eng- 
land. The Boers had effected the isolation of General White 
and his men in Ladysmith by blowing up the two railway bridges 
on the Tugela River at Colenso and Frere, and, promptly on 
learning of the destruction of these bridges, the Natal Govern- 
ment took steps for their rebuilding. The crossing over the 
Tugela at Colenso was designed in five spans of 105 feet each, 
and the crossing at Frere of two spans of the same length. The 
call for the utmost haste in construction was imperative, and 
tenders were invited, both in England and America. The con- 
tract was awarded to The Patent Shaft and Axletree Company 
upon its undertaking to deliver the first span in six weeks from 
the day of the 
contract. The 
order was given 
on the 2 ist of 
December, 1899, 
and the first span 
was finished on 
the ijth of Janu- 
ary, 1900, or in 
nineteen working 

days. When the Norvals Pont Bridge. 

order was received, nothing was in stock at the company's works 
from which the structural steel was rolled, yet at five o'clock on 
the afternoon of the day of the order 100 tons had been rolled at 
the company's works, and tested and approved by the engineer 
of the Natal Government. Each of the spans weighs 105 tons, 
or a ton to the lineal foot of the bridge. There was about 7500 
feet of planing work, and 69,000 rivet holes were drilled in 
each span; yet on January 19, two of the spans had been built 




570 THE DIAMOND MINES OF SOUTH AFRICA 

and work begun on three more, while the material for the whole 
seven for both bridges had been rolled, cut to size, tested, and 
approved. 1 

Besides government railway building, important private lines 
have been constructed for the operation of large mining works, 
and local or suburban traffic. The oldest of these lines was the 
undertaking of the Cape Copper Company, covering a stretch of 
100 miles from Port Nolloth to Ookiep. Some of the grades 
on this line are very notably steep, exceeding any others where 
ordinary steam-engine traction is employed. For this service 
special engines were constructed by Litson & Co. of Leeds, which 
have been working very successfully since October, 1890. On 
Klipfontein mountain there is a rise of 1330 feet in 7-^ miles, and 
in several sections the gradient reaches the extreme of i foot 
in 19. This line was built for the development of the Nama- 
qualand copper mines, one of the most profitable undertakings 
in the Colony. The Cape Copper Company owns most of the 
paying mines, and has been extracting annually about 30,000 
tons of ore, averaging nearly 20 per cent in copper. 

The Indwe Railway Company's line is only second to this, 
with a length of 66^ miles. This line was opened in 1896 to 
reach the Indwe coal mines, and is operated by the Cape Gov- 
ernment as a branch of the Eastern System which it joins at 
Sterkstroom. It was built by the Indwe Company with the 
material assistance of De Beers, which subscribed .75,000 to its 
working capital. The Company owned all its rolling stock, but 
it was operated under the supervision of the Cape Government 
Railway Department. This railway has been lately sold to the 
Cape Government, and is to be extended to Natal. 

It is computed that the lines owned and operated by the 
Cape Government have cost, with their rolling stock, about 
^20,000,000, representing the investment of about ^7200 per 
mile. The capital invested in the Natal lines was ^6,750,000 ; 
showing an outlay of ^1 5,000 per mile. The 777 miles of 

1 The Engineer, London, England, February, 1900. The Scientific American, 
February 24, 1900. 



AN UPLIFTING POWER 



57 1 



Netherlands Company cost nearly 9,000,000, or .13,500 per 
mile. This gives a total of "38,000,000 for the construction of 
3500 miles of railway, not including lines owned and operated 
on private account. With all lines included, it is estimated that 
there is a total outlay of $6 per head of the white population 
of the country, which does not average more than 163 to the 
mile of railway opened. 

In 1896 the earnings of 
the Cape Government rail- 
ways came to something over 
^4,000,000, of the Natal rail- 
ways 1,000,000, and of the 
Netherlands Company nearly 
.3,000,000. The net profit 
after paying interest on capital 
in the Cape was .1,221,675 > m 
Natal "464,762 ; and in the 
Transvaal ,1,328,424, making 
a total of over ^3, 000,000, 
not including the Free State 
share of profit, which for 1896 
was ^289,553. 

Five extensions were au- 
thorized by the Volksraad reso- 
lution of the Free State in 
October, 1896. One line 
through Fauresmith was to serve the diamond mines of Jagers- 
fontein and Kofiyfontein and place them in direct communication 
with the coast ports. In 1898 the Free State decided to build a 
railway by concession from Bloemfontein to Kimberley, and to 
extend the Springfontein-Fauresmith line to join the Bloemfon- 
tein-Kimberley line at a point near Petrusburg. The Springfon- 
tein-Fauresmith line forms a direct route between East London 
and Kimberley, shortening the present route by 100 miles, mak- 
ing East London 40 miles nearer to Kimberley than Port Eliza- 
beth. The Bloemfontein-Kimberley line will reduce the present 




His Excellency Sir Walter Hely Hutchinsen, 
Late Governor of Natal. 



572 THE DIAMOND MINES OF SOUTH AFRICA 

distance by rail between Kimberley and Johannesburg and Cape 
Town. The war has, for the time being, stopped this work. 

The Natal Government is also proceeding with the construc- 
tion of north and south coast lines : one through Verulam to 
Zululand, and the other to the Cape border, where it will connect 
with the extension of the Storkstroom-Indwe line. 

Twenty-five years ago only 781 miles of telegraph were open 
in all of South Africa. A message of twenty words from Cape 
Town to East London cost ijs. 6d. At the outbreak of the late 
war 19,000 miles of wire were working in Cape Colony, and 
probably 10,000 miles in other states and colonies. The march 
of the telegraph through South Africa will be later detailed. 

In addition to the railway and telegraph, several thousand 
miles of excellent roads have been made, and every river of mag- 
nitude has been spanned by substantial bridges. The great 
Zwarte Berg Pass, which rises 3400 feet in eleven miles from 
base to summit, is one of the finest monuments of road con- 
struction to be seen in any country. 

At every port the shipping accommodations have been ex- 
tended and improved, and approaches to the coast have been 
made safer by construction of numerous lighthouses. 

The impulse given by the Diamond Fields development for 
prospecting for mineral deposits of all kinds led to the discovery 
of the mines of Lydenburg, De Kaap, and the Rand. In the 
year preceding the discovery of diamonds Thomas Baines had 
led a party from Durban to prospect for gold in Matabeleland, 
and secured a concession from Lobengula in April, 1870, to 
dig for gold in the district between the Gwelo and Ganyona 
rivers. But Baines's party found no largely promising deposits, 
and without the excitement of the rush to Kimberley there would 
hardly have been any considerable and determined effort to push 
prospecting far beyond the Vaal. Luckily, shortly after the 
rush to the Diamond Fields in 1871, reef gold was found by pros- 
pectors at Eersteling and Marabastad, and, two years later, gold 
placers were discovered about thirty-three miles east of Lyden- 
burg, at Pilgrim's Rest. 




AN UPLIFTING POWER 573 

These discoveries were greatly magnified in the fever of 
speculation excited by the opening of the diamond beds, and 
companies were formed in Natal and England to develop these 
gold-fields, while daring adventurers pushed still farther on, to 
the region north of the Limpopo, seeking the traces of the 
ancient mining works that were known to exist. Upon the 
report of the discovery at Lydenburg some fifteen hundred pros- 
pectors flocked to this field, and a year or two later gold was 
found in the Kaap Valley, fifty miles south of Lydenburg. The 
returns from the placers were hardly tempting enough to hold 
the gold seekers, and conflicts with the natives, followed by the 
outbreak of the war with the South African Republic in 1880, 
were further discouraging to any development in this region. 
After the war the exactions imposed by the South African Re- 
public upon the prosecution of mining in the Lydenburg district 
were a check to outside prospecting. 

In 1882 an Australian digger, Charles Durnin, found some 
very rich patches of gold-bearing ground on the Kantoor plateau 
in the Kaap Valley, and the rush to the Duivels Kantoor and 
Moodies brought to pass the first considerable undertaking of 
gold quartz mining in South Africa. Some gold mines showing 
great richness of ore were soon developed in this district, and 
the bustling mining town of Barberton marked the centre of a 
field which was thought to be of marvellous promise. Unfortu- 
nately the booming of the district ran to a pitch of insane and 
fraudulent speculation that was greatly damaging to the reputa- 
tion of this field of investment, and gold mining undertakings 
in South Africa would commonly have been reckoned as 
"bubbles," had it not been for the uncovering, at this juncture, 
of the astonishing riches of the Rand. 

Nearly twenty years before, the famous elephant hunter, 
H. Hartley, after marking the gold-bearing ground in Matabele- 
land and the region of the Zambesi, made his home on a farm 
in the Witwatersrand, unconsciously settling on the face of 
deposits of gold far more marvellous than any tradition of King 
Solomon's mines. Hartley died without any vision of the treas- 



574 THE DIAMOND MINES OF SOUTH AFRICA 

ures over which he and others were tramping day after day. 
It was soon noticed by roving prospectors, and by settlers in 
the district, that there was gold-bearing sand in the beds of the 
little river and creeks rising in the Witwatersrand, but no note- 
worthy search for gold was attempted until an Australian mining 
man, Armfield, a reputed expert, was sent to prospect in this 
region during the British occupation from 1876 to 1880. He 
made some tests of quartz ledges on a farm adjoining Paarde- 
kraal, but found nothing of value. 

The credit of the first important revelation of gold in the 
district undoubtedly belongs to Mr. Fred Struben, who had 
given his earnest attention to the gold developments in the 
Transvaal, and who prospected the Witwatersberg district in 
1883, and found traces of gold in creeks and reefs, as well as 
ancient workings for copper. In the following year his elder 
brother, Mr. H. W. Struben, purchased two small farms on the 
northwestern end of Witwatersrand, and both the Strubens 
continued their prospecting energetically during the year. In 
the summer of 1884 a gold-bearing vein or reef was discovered 
and traced for several miles by Fred Struben. Ore shoots and 
pockets were found which assayed over one thousand ounces of 
gold and silver to the ton. The rich ledge, named the Confi- 
dence Reef, was supposed to be of prime importance ; and the 
Strubens erected a five-stamp battery on the ground to crush 
the ore of this and neighboring ledges. Several samples of the 
ore were tested in the stamp mill, but the best ore yielded only 
eight pennyweight to the ton. Work on the Confidence Reef 
was greatly disappointing, for the gold-bearing rock was soon 
proven to be a small deposit. 

In August, 1885, I visited a small mine called Kromdraai, 
situated about twenty miles, in a northwesterly direction, from 
the present site of Johannesburg, but at a much lower elevation, 
near the old Pretoria and Kimberley wagon road. A small reef 
of gold-bearing rock was being mined, and the ore crushed in a 
little mill in the immediate vicinity. I also spent a few days 
looking over the Confidence Reef, with the Strubens, who were 



AN UPLIFTING POWER 



575 




Entrance Groote Schuur. 



at that time the most enthusiastic and energetic prospectors, as 
well as the most enlightened and progressive men, in the Trans- 
vaal. Mr. Henry Struben owned large estates near Pretoria. 

The Strubens had spent over i 1,000 in their mining opera- 
tions in the Witwatersrand, and their venture seemed a losing 



576 THE DIAMOND MINES OF SOUTH AFRICA. 

one, when, in the spring of 1886, one of their employees, Walker, 
found the rich reef, now known as the Main Reef Leader, on 
the farm Langlaagte, about two miles west of the present Johan- 
nesburg. In July the first sample of the conglomerate from the 
reefs on the Langlaagte farm was panned in Kimberley. The 
showing was so remarkable that Mr. J. B. Robinson, backed by 
Mr. Alfred Beit, who saw the panning, started on the following 
day for the Rand. The Kimberley-Pretoria coach road ran 
through Potchefstroom, and thence northeast, leaving the little 
pioneer town called " Ferreira's Camp " (now Johannesburg) 
some fifteen or twenty miles to the east ; but Robinson drove in 
a cart from Potchefstroom to the little sprawling camp that was 
the first sprout of Johannesburg. Within a day or two after his 
arrival he bought the Langlaagte farm for ^7000. Scoffers, 
who posed as experts, told him bluntly that " a fool and his 
money were soon parted " ; but he did not take heed of their 
gibes, and, before the end of the year, bought the whole of the 
ground comprised in the holdings of the " Robinson " Company 
for ^13,100. 

Messrs. Rhodes, Porges, Beit, and other enterprising men 
of Kimberley shortly followed Robinson in the pioneer work of 
the Rand. In January, 1897, tne development work on the 
Robinson mine consisted of a hole in the ground, fifty feet deep, 
which was full of water. Robinson, who had been somewhat 
unfortunate on the Diamond Fields, went " nap " on the Gold 
Fields, and the rivalry between him and Rhodes was very keen. 
One story, with some foundation of fact at least, will show this. 
While Rhodes was trying to buy a farm from the Dutch owner, 
and they were parleying in the orchard, Rhodes conversing in 
English and bad Dutch, and the Boer in Dutch and bad English, 
Robinson arrived on the ground. He went direct to the farm- 
house, and at once opened negotiations for the purchase of the 
farm with the Boer's vrouw. His familiarity with the " Taal," 
South African Dutch, was a telling advantage in his competition 
with Rhodes, and he reckoned shrewdly that the wife would jump 
at a bargain more quickly than the husband. So he slapped 



AN UPLIFTING POWER 



577 



a handful of golden sovereigns on the table, saying smartly, 
" Those are for you." The old vrouw clutched greedily at the 
gold and called shrilly to her husband to come to the house. He 
obeyed the call dutifully, and when he entered the door he found 
that his wife had already sold the farm to Robinson. Even a hen- 
pecked man might have grumbled at such a sale, but when the 
simple Boer saw the heap of glittering sovereigns on the table, 




Groote Schuur, after its Destruction by Fire. 

he could not hold out stubbornly against a man who had so 
kindly presented his vrouw with so great " a mark of respect." 
While Rhodes stood in the orchard, Robinson got the farm. 

In the early rush to the Rand, farms and mines were bought, 
not so much for any phenomenal richness, as for the fact that 
they showed more gold distributed over a greater stretch of 
country than had ever been disclosed in South Africa. The first 
two or three years were very disappointing, for the total output 
did not cover the taxes levied upon the mines by the Govern- 
ment. A large percentage of the gold was lost in working the 



2 P 



578 THE DIAMOND MINES OF SOUTH AFRICA 

ores, for the precious metal was so extremely minute that it 
floated away with the water, and, at no considerable depth, a 
portion of the gold was held in the pyrites, and could not be 
recovered by means of the ordinary process of amalgamation. 
Some other process was needed that would save the minutely 
fine gold which became suspended in the water owing to the 
attachment of globules of air. When the Rand was discovered, 
no such process had been developed beyond the experimental 
stage. MacArthur and Forrest, of Glasgow, were experiment- 
ing with a solution of cyanide of potassium, which was known 
to be a solvent of gold. They found that the ores from the 
Rand readily yielded their gold when treated by this process, 
which soon came into general use. This was the saving of the 
Rand, for without such treatment only a few of the richer mines 
would to-day be paying properties. 

A little more than a year after Robinson bought properties 
on Witwatersrand, the despised " cabbage field" of the Lang- 
laagte farm was floated with a capital of ^450,000, and yielded 
^950,000 in gold in the next five years, with a profit of nearly 
seventy-five per cent in dividends on the par value of the capital 
stock. The holdings of the Robinson Company, in the same 
time, produced over ^i, 400,000 in gold and paid ,570,937 los. 
in dividends to shareholders. 

By the discovery of the diamond mines in Griqualand West, 
a product ranging over ^80,000,000 in value in less than thirty 
years had been added to the meagre output of South Africa, and 
the gold mines of the Witwatersrand began, about fourteen years 
ago, to swell this great exhibit of the mineral riches of the land 
by the addition of gold already aggregating over ^70,000,000. 

The annual flow from the diamond mines has averaged, for 
years, over ,4,000,000 in value and the Rand has greatly out- 
stripped even this rich showing. Prior to the discovery of dia- 
monds, the total tally of South African exports and imports 
combined was not ^6,000,000 in value. In 1898 it was nearly 
^50,000,000, and, of the total exports, eighty per cent were 
mineral products. 



AN UPLIFTING POWER 



579 




Pioneers Trekking in Mashonaland. 

With this general survey it is now practicable to trace with 
more clearness the essential and special services rendered in this 
grand development by Rhodes and his associates in De Beers 
Consolidated Mines and other organizations. Viewing, as he 
did, the control of the Diamond Fields very largely as the inter- 
mediary step toward the attainment of an aim far grander, the 
consolidation of the chief diamond mining properties had hardly 
been effected when Rhodes took action swiftly to extend and 
intrench the range of British influence north of the Transvaal 
by obtaining the concession of the mineral rights in Lobengula's 
kingdom of Matabeleland, through the adroit agency of Messrs. 
Charles D. Rudd, Rochefort Maguire, and Frank Thompson, in 
return for an annuity of ^ 1200 and a coveted stock of rifles and 
ammunition. Lobengula made the grant which gave to Rhodes 
the needed nucleus for the creation of the grand exploring and 
developing agency which he pressed for incorporation as the 
British South Africa Company. 

There was a natural hesitancy on the part of the public in 



580 THE DIAMOND MINES OF SOUTH AFRICA 



supporting this scheme at the outset, for the Matabele conces- 
sion seemed to investors at large little more solid than a moonlit 
cloud bank, and even venturesome speculators shrank from buy- 
ing shares in a prospecting license in a country held by savage 
blacks, trained in the school of Chaka to pillage and murder. 
But this incredulity was anticipated by Rhodes, and a solid back- 
ing was given to the enterprise by the subscription of De Beers 
Consolidated Mines for more than ^200,000 of the working 
capital. This was a demonstration of good faith and practical 
intent so convincing that the British Government granted a 
charter formally to the new company in October, 1889, and it 
has since been popularly known as the Chartered Company. 
The government was reluctant to extend the working scope of 
the charter north of the Zambesi, but Rhodes's aim was not pent 

up in Matabeleland, or Ma- 
shonaland, and by his forceful 
representations the British 
South Africa Company was 
left unrestricted in its range to 
the north, as far as it could 
advance without infringing on 
other concessions, or entering 
territory acquired by Germany 
or other nations of Europe. 

There seemed, at first, 
some likelihood of competi- 
tion and possible conflict of 
interests in the race of exten- 
sion with another adventurous 
association, that applied for a 
charter as the African Lakes Company. But the risk was fore- 
stalled by Rhodes's foresight and promptness of action. The 
promoters of the African Lakes Company had spent all the capi- 
tal they could raise, and were so dangerously near the verge of 
collapse that they welcomed the helping hands of Rhodes and his 
friends without much quibbling over the terms exacted. At once 




Khama, a Noted Chief whose Country lies just 
South of Rhodesia. 



AN UPLIFTING POWER 



581 



^20,000 were subscribed by the organizers of the British South 
Africa Company to float the African Lakes Company, and a fur- 
ther subscription of ^9000 a year was pledged in return for the 
right under certain conditions of merging the subsidized com- 
pany in the British South Africa Company. 

Then, with his unhampered charter and its range cleared to 
the source of the Nile, Rhodes was ready, like Davy Crockett, 




Khama's Town. 



to go ahead. After consulting with Frank Selous, the famous 
African hunter, and others familiar with the field, he pitched 
upon Mashonaland as the first base of operations. Dr. L. S. 
Jameson was deputed to go to Bulawayo and get Lobengula's 
express license for this undertaking. The envoy made all pos- 
sible haste in his mission, and won the king's favor so quickly 
by his tactful bearing that the entry to Mashonaland was con- 
ceded. Rhodes lost no time in taking advantage of this oppor- 
tunity. A force of five hundred armed men were enlisted under 
the chartered right to an adequate "police," and two hundred 
pioneers were hired to make a passable wagon road to Mashona- 
land. Colonel Pennyfather was placed in command. 

Meanwhile, the fickle Lobengula changed his mind when 
Dr. Jameson was no longer by his side to persuade him, and sent 



582 THE DIAMOND MINES OF SOUTH AFRICA 

a message to the expedition, forbidding the road-making. The 
messenger of the king met the British at Tuli, but the men 
picked by Rhodes were not of a temper to be checked or fright- 
ened away, and the road was pushed ahead as fast as possible 
through the thick brush and woods of the lowland, where the 
peril from attack was most to be dreaded. Dr. Jameson rode 
in the van with forty of the best mounted men as an advance 




Khama's Hut. 



guard. Selous led the pioneers and marked the roadway. Fin- 
ally, on the 1 3th of August, 1890, the road-makers came to the 
great plateau of Mashonaland, through an easy mountain pass, 
and a heavy weight was lifted from the minds of the leaders, for 
on this open plateau hostile attacks were no longer to be dreaded, 
and a few hundred well armed and mounted men might well defy 
a horde of marching Matabeles. It is probable that this daring 
advance would not have been made unmolested, if Lobengula's 
attention had not been artfully distracted by a feint of entry in 
another quarter made by a body of Bechuanaland police on the. 



AN UPLIFTING POWER 



583 



southwest border of Matabeleland. Thus was the first grip of 
civilization secured on the rich territory which now bears fitly 
the name of" Rhodesia," in lasting commemoration of the grand 
foresight and enterprise of its redeemer from barbarism, Cecil 
John Rhodes. 

No sooner was this entry effected than Rhodes's untiring 
energy sought further extensions of British control. By treaty 
with the native chief, Umtasa, the neighboring Manica was 
brought under the same protecting power as Mashonaland, and 




Matabele Women, Carrying Water. 

a footing was gained with the like expedition in the native prov- 
ince of Gazaland. It was obvious that no extended develop- 
ment of the resources of this territory or stable colonization 
could be effected without railway connection with the Cape, and 
Rhodes at once undertook the provision of capital for the es- 
sential extension of the Transcontinental Railway through Ma- 
feking to Mashonaland. He raised the money required, besides 
drawing heavily upon his private fortune, at the same time, for 
the Beira Railway extension. He contributed also four-fifths 
of the capital of the Transcontinental Telegraph, and, all this 



584 THE DIAMOND MINES OF SOUTH AFRICA 




A Group of Visitors at Groote Schuur. 

while, bearing in great part the extraordinary expenses, amount- 
ing to ^2 50,000 annually for the first two years, of the develop- 
ment of the undertaking of the British South Africa Company 
in Mashonaland. 

Fortunately, by the extraordinary executive ability of Dr. 
Jameson, who was appointed Administrator for the Chartered 
Company in 1891, the immense outlay required of the company 
was reduced to only ^30,000 annually. The thriving town of 
Victoria was founded and the settlement of the country was most 
energetically pressed in spite of every obstacle. But when the 
way for the profitable advance of the company's operations 
seemed to be clearing, its Colony was menaced in 1893 with utter 
destruction by the attack of the fierce Matabeles. 

Lobengula had viewed the entry of the Rhodes expedition 
into the territory north of his kingdom with rising disgust, 
accentuated by his failure to stop it, but it was two years before 
he came to the point of open attack. He had been accustomed, 
all his life long, to regard the district occupied by his neighbors. 



AN UPLIFTING POWER 



585 



the weak and unwarlike Mashonas, as convenient harrying 
ground for his brutal forays. Marauding troops of freebooters 
were constantly harassing the poor Mashonas, and oftentimes 
the king would send his robbing and murdering expeditions to 
scourge the land, just as he sent his impis to take Ugami, to 
despoil and enslave and massacre the Batuwani, and, across 
the Zambesi, to raid the Mashukulumbwe or the Barotse. 

To the sorely persecuted Mashonas the coming of the Eng- 
lish was an assurance of protection which was greatly welcomed, 




Great Kafir Kraal, Mochudi, Matabeleland. 

but even the presence of the bold white men and the unfolding 
of the British flag did not stop the marauding. Dr. Jameson 
protested over and over again to Lobengula, but the king was 
deaf. Finally, in July, 1893, parties of the Matabeles pushed 
their ferocious raids contemptuously up to the very bounds of 
the township of Victoria, and the English could not look on 
unmoved. Then Dr. Jameson sent a squad of police to warn 
off the marauders. The Matabele insolently fired on the guard, 
and the police charged and drove them flying. 



586 THE DIAMOND MINES OF SOUTH AFRICA 

This wholly rightful rebuff upset the temper of Lobengula, 
who was stuffed with barbaric conceit. His impis began pour- 
ing over the border, and the infant Colony was threatened with 
extinction. The menace was met by a heroic response. There 
were only a handful of police at the time in Mashonaland, but 
the settlers were men who could defend themselves. 

It was judged best to meet the roving assaults by a direct 
counter attack on Lobengula's stronghold, his capital of Bula- 
wayo. The Chartered Company's funds were drained out ; but 
Rhodes, as ever, rose to the occasion, and raised the money im- 
peratively needed to arm and conduct the little force that was to 
make the daring venture into the heart of the most savage and 
warlike province in South Africa. With this backing Dr. Jame- 
son raised a force of about nine hundred men, and, placing himself 
at their head, as Commander-in-chief, marched on Bulawayo. 
Just after crossing the Shangani River, his little army was at- 
tacked by the Matabele in force, but beat off their assailants. 
With the encouragement of this success the English pushed 
on to the Imbesi. 

Here they were attacked again in the old Zulu fashion by 
desperate charges of seven thousand frantic blacks in rank after 
rank of impis upon the well-prepared English camp. It was 
a fierce fight, but the issue did not hang long in doubt. The 
Matabele were as dashing and reckless as the impis that had 
fallen like breakers of surf on the laagers of the Boers during 
their " Great Trek." But they were overmatched by the cease- 
less belching of machine guns and repeating rifles, mowing them 
down swath by swath when they charged within close range. At 
last they broke and fled, and Dr. Jameson's little army marched 
on to Bulawayo, which was entered without further fighting, for 
the disheartened Lobengula abandoned his capital. 

The British pursued him hotly, for it was highly important 
to put a decisive end to the war by his capture before the 
advance of the rainy season. Unfortunately this pursuit was 
too daringly pressed. Major Wilson, with a little force of less 
than forty mounted men, nearly plucked Lobengula out of the 



AN UPLIFTING POWER 



57 




5 88 THE DIAMOND MINES OF SOUTH AFRICA 



midst of his retreating impis. His impetuous rush reached the 
cart that carried the king; but the desperate Matabele flung 
themselves upon this little troop in such masses that its advance 
was checked. Messages for help were sent to the main troop 
in pursuit, under Major Forbes, but succor was cut off by the 
rising of the Shangani River. Flight would have saved most of 
them, but Wilson and his men who were able to ride off scorned 
to abandon their wounded comrades. So, hard pressed by the 
Matabele on all sides, they made a barrier of their horses, living 

and dead, and 
held their ground 
until their last 
cartridge was 
fired. Then they 
stood up defiantly 
and fought hand 
to hand until the 
last man was cut 
down and tram- 
pled under foot 
in the crush of 
the savage blacks. 
The troop under 
Major Forbes was 
forced to retreat, 
and suffered much privation before it was met by a relief party 
headed by Rhodes, who rode out from Bulawayo. 

The loss of Major Wilson and his gallant men was deeply 
mourned ; but the campaign as a whole was a most brilliant 
success. Lobengula's power was completely broken, his impis 
scattered, and he soon afterward died a fugitive. The royal 
city of Bulawayo was made the capital of Rhodesia, the province 
of the Chartered Company, and Dr. Jameson took his seat there 
as Administrator. The rich mineral ground near Bulawayo soon 
attracted a considerable influx and made a rising town, which in 
less than three years boasted of its banks, clubs, newspapers, 




Major Wilson's Grave. 



AN UPLIFTING POWER 589 

electric lighting, and water-works. The brave colonists who 
made up the force of " Jameson's Volunteers " were disbanded, 
and began to prospect for gold and pick out farms in the new 
province. 

With the fall of Lobengula, a standing menace to the march 
of settlement was removed, and the attractions of Rhodesia 
began to come out in a brighter light. It was the settled pur- 
pose of the Chartered Company, from the outset, to do every- 
thing that was feasible to encourage investigation and the taking 
up of farms by honest and thrifty colonists. This was regarded, 
by Rhodes at least, as transcending in importance even the 
development of the mineral riches of the country, though the 
latter was naturally the chief object with most investors. Par- 
ticular pains has been taken, in directing the colonization, to 
harmonize relations between the men of different races and 
nations, and to draw as closely as possible all together in a com- 
mon bond of union as Africanders. 

Considerate and elevating treatment of the natives has also 
been a notable feature of the determination and policy of the 
Chartered Company. The relief of Mashonaland from the 
ferocious forays of the Matabele was a memorable service 
which will be credited at the outset to this company. It has 
further given to all within its jurisdiction the fullest protection of 
English law, and safeguarded all working in service from abus- 
ive treatment by their employes, prohibiting the use of the lash, 
and enforcing other humane regulations. The sale of liquor to 
the natives is forbidden by stringent laws, and the most discred- 
itable and demoralizing influence in South Africa is barred out 
of Rhodesia, at least. 

To determine the extent of arable and pasture lands, deputa- 
tions of experienced farmers were appointed to inspect and report 
by public meetings in the Cape Colony and Orange Free State. 
Their examination of less than half the area of Mashonaland and 
Matabeleland reached the conclusion that at least 40,000 square 
miles were well adapted for colonizing purposes. It may further 
be noted that highly favorable reports of the agricultural and 



590 THE DIAMOND MINES OF SOUTH AFRICA 

mineral resources in the vast territories of the British South 
Africa Company, north of the Zambesi, have been furnished by 
Joseph Thomson, Alfred Sharpe, and other well-known ex- 
plorers. Actual experience and the medical officers' reports have 
shown that the climate is not unhealthy for any white man who 
will avoid undue exposure and observe a few simple precautions. 1 




Victoria Falls, Zambesi River. 

The advance of immigration and development has been 
remarkable in view of existing conditions. There were inevitable 
hardships and discouragement to check the first rush of gold 
seekers. The gold-fields were only slightly explored and lay far 
from any base of supply. There was lack of resources and 
means of communication to develop even the most promising 

1 " Minutes of Progress in Mashonaland," by the Secretary of the British South 
Africa Company. "The Ruined Cities of Mashonaland," pp. 405-412. 



AN UPLIFTING POWER 



openings. Yet, in spite of all obstacles, prospecting has been 
pushed far enough to show the range of gold-bearing ground 
and ledges for hundreds of miles. Convincing evidence of the 
mineral richness of the country is given to the extent of the 
ancient workings that have been traced through Mashonaland 
far beyond the southern end of Lake Tanganyika. There can 
be no doubt that an enormous amount of gold has been taken 




Victoria Falls. 



from this region both by placer washing and quartz mining. 
There are no other ancient workings, on the face of the Old 
World at least, of like extent, and this undeniable evidence 
weighs heavily for the contention that the flow of gold from this 
source was the main supply, for centuries, of Arabia and Asia 
Minor. In view of the superiority of modern appliances for 
mining and the extraction of gold, it would seem, at least, 
probable that the yield of this territory may, in time, be large. 
There are apparently well-verified reports also of the dis- 



592 THE DIAMOND MINES OF SOUTH AFRICA 



covery of extensive copper and iron deposits in North Rhodesia 
and in the region lying along the western shores of Lake Tan- 
ganyika. The missionaries of the Roman Catholic society 



/ UKITlVlI .pi 
/ CE.VrilAL I V.\ 

i >? Iw 

iiiuri V; 

{ KULANTYfTE 



,,^>i4. iro ^VR* 

5^A^r ; s*$$ r 

--t. Zv-voHh&B in^Si^. .. n 




A MAP 

TO ILLUSTRATE 

THE RHODESIA RAILWAY SYSTEM 

COMPLETED, UNDER CONSTRUCTION, 
AND PROJECTED. 

1902 
SCALE OF MILES 



Riilwiji completed 

" under conatructloa* 



known as the White Fathers have long been at work on the 
shores of Lake Tanganyika, and a report of their explorations 
has been published lately in Petermanns Mitteilungen. It is 



AN UPLIFTING POWER 



593 



noted that great quantities of iron ore have been found along the 
banks of the rivers flowing into Tanganyika, particularly along 
the Lufuko and Miobosi. 

The wide-ranging Marungu district is said also to be exceed- 
ingly rich in copper ores ; and the copper areas, better withstand- 
ing denudation than the surrounding country, are reported to 
stand comparatively high 
above the general level and 
to be easily recognizable. 
Agents of the Chartered South 
Africa Company have also 
reported the discovery of a 
rich copper field, estimated 
to cover 40 square miles, in 
north Rhodesia. This field 
lies about 150 miles north of 
Victoria Falls, near the Congo 
Free State, and runs over the 
border. De Beers Company 
has already taken an active 
part in the development of the 
copper mines of Namaqua- 
land, and the new field may 
prove to be of even greater 
importance. 

The rapid extension of the 
railway lines of the Bechuana- 
land Railway Company from 
Vryburg to Bulawayo was mainly due to the aid given by Mr. 
Rhodes and his associates in the Chartered Company. This 
line reached Bulawayo, a total distance of 1360.4 miles from 
Cape Town, in 1897, an d has since been extended northward 
about 30 miles. The war has interrupted this work during the 
past two and a half years. The main line north, it is expected, 
will reach the enormous coal beds at Wankie, 200 miles north 
of Bulawayo, in about eighteen months, and will be pushed on 




Victoria Falls. 



594 THE DIAMOND MINES OF SOUTH AFRICA 




AN UPLIFTING POWER 595 

to the Victoria Falls on the Zambesi and thence north as rapidly 
as possible to tap the rich copper districts. 

The telegraph line running ahead of the railway was carried 
across Rhodesia and reached Lake Tanganyika, from which it is 
fast extending to Uganda, so that Cape Town and Cairo will soon 
be in direct overland telegraphic communication. The inspira- 
tion of this work of the Transcontinental Telegraph Company 
was due to Rhodes, and the greater part of the capital needed to 
extend it was contributed by him personally. The postal ser- 
vice already effected is as remarkable as the telegraph. Even 
from points hundreds of miles beyond Bulawayo, which, eight 
years ae;o, was the heart of savage Matabeleland, the pioneer can 
send a letter home to England for twopence halfpenny, and the 
settler on the remote shores of Lake Moero can get by mail a 
pound package of tea from Liverpool at the cost of a shilling. 

Already the Chartered Company has carried the work of 
exploration and expanding control to Lakes Tanganyika, Moero, 
and Nyassa, and made treaties with the native chiefs north of 
Rhodesia, as well as with Lewanika, king of the Barotse, to the 
west. The range of British influence and civilizing advance 
now reaches to the heart of Africa from the south, embracing all 
the country not within German control in the west, and the Portu- 
guese domain in the east. The grand aim of Rhodes has been 
swiftly advanced in realization even beyond sanguine expectation. 

Another undertaking, less far-reaching and impressive in scope, 
but of evident material importance to the development of the in- 
dustrial resources of South Africa, was liberally and energetically 
supported and advanced by De Beers Mines. The cost of import- 
ing coal was from the outset, and still continues to be, a crippling 
handicap upon the advance of the mining and manufacturing in- 
dustries of the South African Colonies. Persistent searches for 
coal deposits throughout the country were made, but no coal 
seams of high quality were uncovered. The best apparent pros- 
pect for opening deposits that might compete in the market with 
imported coals was shown in the Stormberg and neighboring hilly 
districts lying between Queenstown and the Orange River. 



596 THE DIAMOND MINES OF SOUTH AFRICA 




Summer House at Groote Schuur. 



Progressive colonists developed the outcrops in the Storm- 
berg district, and in the face of grave discouragements opened 
seams of importance in the Molteno and Cyphergat mines, but 
it was impracticable to work these mines with any prospect of 
profit until railway communication was opened from Stormberg 
Junction via Steynsburg to Middelburg, connecting the East 
London and the Midland or Port Elizabeth lines. The possi- 
bility of supply from this district was immediately grasped by 
De Beers Company, both for the sake of an eventual saving in 
the cost of its fuel, and the public-spirited object of cooperating, 
so far as was feasible, in the development of a resource of such 
importance to the colonies. 

The Stormberg coal was so mixed with shale that even the 
shipping coal after sorting held about one-third waste, which 
clogged the furnaces. But special grates were designed to burn 
this coal, and by this resort it was practicable to use a supply 
from this field at the diamond mines. De Beers Company was 
soon taking by contract practically the entire product of the Storm- 
berg seams at a price of about 2OJ. per ton at the shipping point. 

Not long after the opening of the Stormberg mines, coal 
seams of much greater width and promise were discovered at 
Indwe, a point about seventy miles east from Molteno and 
Cyphergat. Here the prospective returns from energetic devel- 



AN UPLIFTING POWER 



597 



opment were really very bright, but, to market the coal, the con- 
struction of an expensive railway line from Indwe to the East 
London or Eastern systems was indispensable. In spite of the 
unwearied and cogent representations of Colonel Schermbrucker 
and his associates in control of the Indwe field, the Cape Gov- 
ernment was reluctant to defray the cost of building this line. 
The scheme was a dragging one for years, until De Beers Com- 
pany came forward 
with a subscription 
of ^7 5,000 to the 
shares of the Indwe 
Railway Collieries 
and Land Company, 
organized to extend 
the necessary railway 
lines and operate the 
mines. 

In view of this 
essential backing of 
capital, coupled with 
the cogent appeals of 
Rhodes and his as- 
sociates, the Cape 
Government was 
moved to contribute 
a grant of ^50,000 
toward the expense 
of construction, with 
an additional allow- 
ance of 50,000 acres of land, worth about one pound an acre. 
Then a line of sixty-six miles was laid at half the rate per mile that 
was paid for building the lines under Government Administration, 
and the mines were opened very successfully. It was supposed 
by the projectors of the scheme at the outset that the main busi- 
ness of the company would be the supply of coal for steamship 
use at East London ; but it was soon demonstrated, upon the com- 




' Stoep," Groote Schuur. 



598 THE DIAMOND MINES OF SOUTH AFRICA 

pletion of the railway, that De Beers Company was the principal 
customer, consuming about 5500 tons of the average monthly 
production of 12,000 tons. This coal supply was delivered to 
De Beers by agreement for i$s. per ton at Sterkstroom, the 
point of junction of the Indwe and Eastern system lines. In 
spite of the inferior quality of the coal, compared with Welsh 
coal, the South African coal at this price was a good bargain for 
De Beers, and the very profitable record of the Indwe Company 
proves that the interests of its shareholders were not sacrificed in 
making the bargain. The mines of the Stormberg district are 
still continuous producers, and supply about 1000 tons monthly 
to the mines at Kimberley not under control of De Beers 
Company. 

Coal mining in the Orange Free State has not been carried 
on very energetically on account of the distance of the coal meas- 
ures from the existing railways. But the developments in this 
field are already promising, and the Kroonstad Coal Company, 
in particular, has opened up a bed of very good coal. A rail- 
way is in course of construction from the main Free State line 
to the Kroonstad coal fields. When this line is completed these 
mines will be in a position to compete with any others, and if 
the long-promised line is constructed from Branford or Bloem- 
fontein to Kimberley, Kroonstad coal can be delivered at the 
diamond mines cheaper than any other coal yet discovered. 
Beyond these undertakings is the opening of the promising coal 
mines in Natal to which De Beers Company has liberally con- 
tributed. (See Appendix IV.) 

Other enterprises, too, of public service are worthy of 
mention. De Beers Company is steadily furthering fruit and 
stock farming, and has constructed storage buildings in various 
locations in order to prevent a monopoly of the meat supply 
which was threatening South Africa. It is constructing, also, 
one of the largest dynamite factories in the world, near Cape 
Town, under the able superintendence of Mr. W. R. Quinan. 

Of course Rhodes could not foresee the marching steps of 
this progress in varied lines, but it is none the less certain that 



AN UPLIFTING POWER 



599 



the expansion of the undertakings of De Beers Consolidated 
Mines was the carrying out of his long-cherished aims. It was 
for this chiefly that De Beers Charter was drawn with so free a 
hand. Assured control of the great South African diamond 
mines was the assurance of great wealth, from Rhodes's point 
of view, great power that should be greatly used. His aims 
ranged far beyond 
any personal ex- 
alting. His heart 
was set on the 
making of Greater 
Britain by expan- 
sion and loyal 
federal union. In 
the Dark Conti- 
nent, beyond the 
confines of civili- 
zation, he saw the 
open field for Brit- 
ish occupation and 
development, and 
was unresting till 
it was grasped. 
How great this at- 
tainment was in 
actual stretch of 
territory may best 
be comprehended, 
as the London Times notes, "by any one who will take the 
trouble to contrast the map of Africa as it appeared in 1881, 
when Mr. Rhodes first entered public life, with that which is 
open to his study to-day. At the earlier date, the line of the 
28th degree of south latitude bounded our possessions in South 
Africa ; the later map he will find coloured red right up to the 
shores of Lake Tanganyika within a few degrees of the 
Equator." 




A Corner in Mr. Rhodes's Library. 



6oo THE DIAMOND MINES OF SOUTH AFRICA 

That this annexation has been, and will be, greatly to the 
advantage of the territory and its occupants will not be seriously 
questioned. Its material advance and the security to life and 
property stand already in bright contrast to its barbaric state 
a land which knew only the rudest tillage and was ravaged at 
the whim of savage chiefs. It is too early yet to think of meas- 
uring its resources and probable advances, but enough is known 
to warrant high confidence in its future, with the assurance of 
alert grasp of its openings for immigration and capital. 




Another View of Mr. Rhodes's Library. 

To any eye the gaining of Rhodesia was a long step forward 
toward the attainment of Rhodes's hope of carrying British 
dominion from the Cape to Cairo. But the ordinary observer 
would not mark, as intently as Rhodes did, the force of this 
acquisition in determining the control of South Africa. Seven- 
teen years ago, in addressing his constituents, at Barkly West, 
he declared publicly, as a settled conviction: "I came to the 
conclusion that the key to the (South African) puzzle lay in the 
possession of the Interior, at that time an unknown quantity. 
In a humble way I have been mixed up with the politics of the 



AN UPLIFTING POWER 60 1 

Interior during the last four years, and such politics, I contend, 
will be in future most intimately connected with the settlement 
of the South African Question, for I believe that whatever State 
possesses Bechuanaland and Matabeleland will ultimately possess 
South Africa." It was his view, asserted in repeated conversa- 
tions with Mr. Edward Dicey, that the taking of Rhodesia 
necessitated the creation of a predominant South African Con- 
federacy, which would be brought to pass by the force of cir- 
cumstance. In the interest of South Africa and Great Britain 
Rhodes sought the inclusion of this Confederacy in the British 
Empire. 

It is plainly to be seen that Rhodes's view of the interests 
of South Africa and the drift of his anticipated confederation 
were inevitably antagonistic to the attitude and policy of the 
men controlling the South African Republic. In stretching 
the arm of Great Britain over Mashonaland and Rhodesia, 
Rhodes unquestionably blocked the extension of the Transvaal 
State and the schemes of Kriiger. In this brief marking of 
progress and attainment I would not attempt any measuring of 
responsibility for the collision that finally resulted in the war 
just closed. South Africa is now completely under British Im- 
perial control. Whatever view may be taken of the conflict, 
its practical outcome plainly clears the way for the systematic 
development of this vast territory under liberal colonial insti- 
tutions. 

Cecil John Rhodes did not live to see the ending of the 
contest so long maintained by the unyielding temper of the 
Boers. He died on March 26, 1902, near Cape Town, of 
the disease of the heart which had long clouded his hope of 
life. 

His visionary political projects ran far beyond any exact 
defining or determination of method, but, in the main, " the 
lay of his ideas," to use his own phrasing, is clear. He would 
urge the union of all English-speaking people to dominate the 
world, transform barbarism to civilization, do away with poor 
and hampering government, maintain enduring peace, and pro- 



602 THE DIAMOND MINES OF SOUTH AFRICA 

mote universal progress. His last will and testament has proven 
that the advance of this union was at the core of his heart. The 
image of his fancy may never come into being, but he has, at 
least, done something for an uplifting union in the gathering of 
young scholars, representing all English-speaking people, in the 
ancient mother university, to recall their common inheritance 
and join their hands. 



APPENDICES 



APPENDIX I 

THE MINES BESIEGED 

THE siege of Kimberley was one of the striking episodes of the late 
war. As an interruption to the peaceful progress of diamond mining in 
the South African Fields, it has a place apart from the industrial story. 
Yet no history of the Diamond Fields would be complete without some 




A Boer Commando. 

account of its course, and my personal view may be of interest in the 
possible emphasis of the part taken by De Beers in the maintenance of 
the defence. I would mark, too, precisely how the war affected the work- 
ing of the mines, and tell from my own observation how the call to arms 
made soldiers of men accustomed to the use of drill, pick, and shovel, and 
caused our mechanics to turn their hands to the making of ordnance. 

605 



606 THE DIAMOND MINES OF SOUTH AFRICA 




For some time previous 
to the actual outbreak of 
the war (October 1 1, 
1899), it was apparent to 
us who were living upon 
the border of the Orange 
Free State that both the 
South African Republic and 
the Orange Free State were 
making preparations for 
war with England, and that 
the invasion of the Cape 
Colony was but a matter 
of a short time. These 
preparations had been going 

Lieut. Col. Kekewich and Staff of Imperial Officers. Qn f or man y y ears unt il the 

magazines and arsenals of the Transvaal were filled with the finest 
munitions of war that the works of Schneider at Creusot or of Krupp 
at Essen could produce. The Mauser with which the Boers were armed 
was as good as the small arms of any 
Continental power, and better than 
the Lee-Metford which the British 
brought against them. 

In July, 1899, Major Scott-Tur- 
ner came to Kimberley,. and Lieu- 
tenant Mclnnes, Royal Engineers, 
followed him shortly after. Colonel 
Trotter, R. A., Chief Staff Officer, 
also came to stay a short time. He 
had made a report on the defences 
of Kimberley as early as 1896, and 
an accurate military map had been 
prepared of the town and surround- 
ings. Major O'Meara came later as 
Intelligence Officer. The Imperial 
Government sent these officers to 




Lieut. Col. Robert George Kekewich, Com- 
mandant of Kimberley during the Siege. 



prepare for the defence of Kimberley, and on the I3th of September, 
shortly before the war was declared, there arrived a half regiment of the 
Loyal North Lancashires (infantry), and a battery of Royal Artillery, 



APPENDIX 1 



607 



consisting of six muzzle-loading seven-pounders of obsolete pattern, and 
some Maxims. 

On the 30th of September the Governor of the Cape Colony gave 
his consent to the formation of a Town Guard, " solely for local defence 
in case of attack from without." The radius of the circle in which the 
Town Guard must confine their operations was eight miles, with the 
market square as the centre. Lieutenant Colonel Robert George 
Kekewich was appointed commandant. Lieutenant Colonel Harris, 
V. D., a director of De Beers, was second in command and was placed 
in charge of the Town Guard. Major Peakman, an officer of the local 




Fort on Tailings Heap, Kimberley Mine Floors. 

volunteer force, who had had a considerable amount of experience in the 
Kafir wars, was appointed Staff Officer. On the 4th of October the 
local volunteers, five hundred strong, were called out by the Governor, 
and went into camp. 

On the 5th of October the first serious disturbance of the work at 
the mines occurred. An alarm was sounded at one o'clock in the morn- 
ing of that day, and all the forces in town, including the men working in 
the mines, were called out to do military duty, as it was rumored that 
an attack was contemplated by the Boers, who were massing commandoes 
in the Orange Free State, only a few miles distant. It had been arranged 
that the whistles (sirens), commonly called " hooters," at the various engine 



608 THE DIAMOND MINES OF SOUTH AFRICA 

houses of De Beers Company, should be blown in case an alarm had to 
be given. The first alarm caused great consternation throughout the 
whole town. Men were running, helter-skelter, in the dark, seeking 
their various redoubts, the moving guns and ammunition wagons rattled 
through the streets, and the gardens of the houses were filled with men, 
women, and children, anxiously awaiting some news as to the cause of the 
alarm. The screeching of the hooters was appalling. These sirens, 
which in times of peace could " blow the boilers dry " and not disturb 
the quiet morning slumbers of the dwellers of the Diamond City, had, 
all in a moment, become a nerve-shattering mechanism. In later days 
the roar of the Boer artillery and the bursting of shell all over the town 
did not so frighten the mass of people. The horrifying effect was so last- 
ing, that when work at the mines was resumed after the siege, many people 




One of the many Redoubts, looking East. 

in the town asked me to discontinue the use of the hooters, and, in compli- 
ance with' their wishes, the old whistles were for a time put into service. 

Kimberley, as may be imagined, was quite unprepared for an attack on 
the 5th of October, as war had not been declared. The Intelligence De- 
partment had received some false reports, and those in charge thought it best 
to have every man at his post ; hence the alarm. The proven falsity of 
the reports did not, however, dispel the menace of the situation, and it 
was considered necessary to make better preparations for the defence of 
the town. Our miners were called out to drill during a part of each 
day. Our tailing heaps, which formed natural defensive positions, 
were taken possession of by the military. Strong forts and redoubts were 
constructed on the tops of these heaps, and mines of dynamite were 
laid at their bases. 

Sir Edwin Arnold, writing in the Daily Telegraph, said: "There is 
something singularly picturesque and suggestive in the thought of the 
Diamond City of South Africa being defended by her own waste heaps. 



APPENDIX I 



609 





View from the Conning Tower, looking Northeast. 



Z R 



610 THE DIAMOND MINES OF SOUTH AFRICA 

Since Syracuse was fortified against Nicias with the columns of her own 
white marble temples, and the breaches of Badajoz were filled up with 
the empty wine casks, there has been no such curious use made of local 




The \VuU-r\\ urks Company's Reservoir as a Fort. 

material. Strange, indeed, is the destiny of matter. It may turn out 
that the blue clay will prove more valuable to Mr. Rhodes, to the isolated 
garrison, and to the little city, than all the diamonds she ever dug up." 




Fort Rhodes, on Top of No. 2 Tailings Heap, De Beers Floors. 

Other defensive fortifications were made upon the ground lying 
between the tailing heaps. The labor necessary to do this work was 
drawn from the mines and works. Nearly all the men working in and 
about the mines joined the various military organizations, and slept in the 



APPENDIX I 



611 



forts and redoubts. Owing to this distraction, work at the mines pro- 
ceeded very slowly. 

Rhodes, accompanied by Dr. Smartt, member of the Legislative As- 
sembly, arrived in Kimberley a few days before the investment. He took 




No. 2 Redoubt, near Kimberley Mine. 

up his residence at the Sanatorium. Mr. and Hon. Mrs. Maguire 
arrived a day or two later and were his guests during the siege. Upon 
his arrival at Kimberley, Rhodes realized at once the gravity of the situa- 
tion, both as regards the defence of the town and the food supply. Orders 
for large quantities of provisions were wired to Cape Town, Port Eliza- 




A Barrier on the Road leading to Kenilworth. 



beth and East London, with the hope that we might be able to add to the 
seemingly large stock already on hand but these supplies never arrived. 
The siege of Kimberley commenced on the night of the I4th of Octo- 
ber, a little before ten o'clock, when the wires to the south were cut, the 



612 THE DIAMOND MINES OF SOUTH AFRICA 




The Sanr.torium in Time of Peace. 




The Sanatorium during the Siege, 



APPENDIX I 



613 



wires to the north having been cut about an hour before. The last train 
from the south arrived at Kimberley about 1 1 P.M., bringing several 
truck-loads of supplies which were at Modder River Station, destined for 
the Free State Boers. 

Colonel Kekewich at once issued a proclamation, declaring the 
district in a state of siege. The war had actually begun. The various 
fortifications were made stronger, military organizations were increased 
in numbers, a mounted force of four companies, known as the Kimberley 
Light Horse, was formed, and on all sides there was the greatest activity 




Mounted Camp, Kimberley. 

in making Kimberley a strongly garrisoned town. When all the military 
organizations were completed, the forces at the disposal of Colonel 
Kekewich were as follows : 

The Imperial Garrison consisted of the 23d Company of Royal 
Artillery, one section of the yth Field Company Royal Engineers, and 
four companies of the 1st Royal North Lancashire Regiment. There 
was also a small detachment of the Army Service Corps. The total 
strength of the regulars was about 600 officers and men. Volunteer 
companies had been enrolled from the early days of the Fields, 
and at one time comprised a very considerable force of men, but of late 
years the community had lost nearly all interest in the volunteer service. 



614 THE DIAMOND MINES OF SOUTH AFRICA 

Still the organizations had been kept up, and when the muster roll was 
taken, shortly before the siege, it showed the following numbers : One 
battery, Diamond Fields Artillery, consisting of six seven-pounder muzzle- 
loading guns, with 3 officers and 90 men, in charge of Captain May; the 
Kimberley Regiment (infantry), under Lieutenant Colonel Finlayson, with 
14 officers and 285 men; the Diamond Fields Horse, Major Rodger, 6 
officers and 142 men. The total force of regulars and volunteers was 
about 1 100. 

The Town Guard was organized, and the men were drilled in the 
use of the Lee-Metford rifle. At the beginning of the siege this force 
numbered about 1200 men, but both the volunteer corps and the Town 




Camp of the Royal North Lancashires. 

Guard were soon increased until the total strength of the garrison 
reached 4500 men. This included the Cape Mounted Police, number- 
ing about 360 officers and men, and unmounted police to the number 
of 175. The limit of the defence force was gauged by the number of 
rifles in Kimberley which had been considerably increased during the 
previous year by the importation by local merchants of 1000 rifles and 
six Maxims, together with a considerable amount of ammunition for the 
use of the rifle clubs. 

Our forts and redoubts were in many ways unique and picturesque. 
The waterworks reservoir was surrounded by a huge fortification, made 
of grain and coal sacks filled with soil. The forts on the tailing heaps 
were made with rows of the trays of trucks which in times of peace 



APPENDIX I 



615 



convey the diamond-bearing ground to the floors. The trays were filled 
with tailings, banked up on the outside with the same material, and 
coped with sand bags. Large shelters were made within the forts for 




Officers of the Diamond Fields Horse. 



the protection of the garrisons. As tents were not to be obtained, 
spacious houses with roofs of corrugated iron and sides of canvas were 
constructed as sleeping and eating rooms, and for protection against the 
tropical sun and violent thunderstorms. When the supply of corrugated 




The Diamond Fields Horse at Kenilworth, during Siege of Kimberley. 

iron gave out in town, for even the enormous stock of De Beers did not 
prove equal to the demand, the iron fence which surrounded the race- 
course was taken down and carted to the various fortifications. 



616 THE DIAMOND MINES OF SOUTH AFRICA 




A Group of the Town Guard. 

Barriers were constructed around Kimberley to check any sudden 
attack upon the town. The roads leading from the town were strongly 
guarded and barricaded with barbed-wire entanglements, with mining 
trucks filled with earth, and with camelthorn trees. Of late years the 
outskirts of Kimberley had begun to assume quite a parklike appearance, 
by the growth of young trees from the roots and stumps of those that 
had been cut down during the early days of the Fields. It seemed a 
pity that the little natural beauty which these afforded should be destroyed; 
but the preservation of the town was of first importance, and all the trees 
were cut down and dragged into long lines of fences, where they were 
interlaced with barbed wire, making most formidable barriers. When 




Defence Guns and Maxims massed in the Gardens, 



APPENDIX I 



617 



the siege was over, these fences disappeared, almost in a day, to supply the 
inhabitants with firewood, which had been cut down to the scantiest 
allowance, a week's supply being barely sufficient to do a day's cook- 
ing. The defences were in places supplemented with dynamite mines 
planned by the Royal Engineers, and carried out by the electrical depart- 
ment of De Beers. On one occasion the officer in charge gave instruc- 
tions to put down ten pounds of dynamite every thirty feet, and returning 




Headquarters Staff, Cape Mounted Police. 

later in the day he asked if his instructions had been carried out, and 
received the reply, "Yes, sir, we have put down thirty pounds of dynamite 
every ten feet." 

Premier Mine 

Premier mine occupied a unique position during the siege. It was 
isolated from Kimberley and Beaconsfield, the former town being about 
four miles, and the latter two miles, distant. There is a large, disused 
tailing heap near the mine, on the top of which is a small reservoir, into 
which water from the mine is pumped for distribution to the washing 
plant and floors. Around this reservoir a fort was built and made almost 
impregnable. Large shell-proofs were made for storing supplies and 



618 THE DIAMOND MINES OF SOUTH AFRICA 



ammunition for a local siege, should communication with the Kimberley 
and Beaconsfield defences be cut off. One of the three searchlights 
which De Beers Company uses on their "floor" for preventing theft 
of diamonds by night was placed at this fort. The Boers called these 




Kimberley Waterworks Reservoir, with Royal Artillery. 

searchlights "Rhodes' eyes." About 150 of De Beers employes and 

one hundred regulars, with two seven-pound guns and a Maxim, were 

constantly on duty at this fort. 

The pumping plant which supplied Kimberley was down in the open 

mine. This plant, as well as all the machinery of the mine, was pro- 
tected with sand 
bags. In heaps 
about the mine, and 
in all the buildings 
on the side of the 
mine adjoining the 
Free State, mines 
were laid, with wires 
leading from them 
to the fort. One 
of the powerful elec- 




Premier Mine Fort, Royal Artillery in Action. 



trie searchlights was placed in the fort, and so arranged that it could be 
lowered out of harm's way during the daytime. Connections were made 
between the two sets of boilers and the pumping and electric light plants, 
so that, in case a shell damaged one set, the other could be used. A large 



APPENDIX I 



619 



number of hand grenades filled with 
dynamite, with fuses and detonators 
fixed, were made and kept in the 
magazine. An underground hospital 
for the wounded was constructed. 
In fact, everything necessary was 
done to make this fort independent 
and secure. There was apparent 
need for these precautions, for the 
Boers constructed the most formi- 
dable fort of any about Kimberley 
on a low range of hills about three 
miles distant, where they kept two 
guns and a pom-pom, which they 
fired nearly every day during the 
siege, except Sundays. 

Great credit is due to the man- Premier Mlne Searchli g ht - 

ager of the mine, Mr. J. M. Jones, and to Captain O'Brien, who was 
in charge of the garrison, for the manner in which the defences were 
constructed, and to all who occupied the fort during the long, weary four 





Canvas House erected for Protection from the Sun and Thunderstorms. 

months, for their courage and patience. On several occasions lightning 
struck the wires connecting the mines and exploded them. One explo- 



620 THE DIAMOND MINES OF SOUTH AFRICA 

sion carried away part of the mine compound, and another wrecked the 
end of the large stables. Fortunately no harm came to any of the gar- 
rison or to any of the machinery of the mine. Although it was isolated 
from Kimberley, the Boers never made an attack upon it nor came within 
rifle range. 

A few days after the commencement of hostilities the Boers took pos- 
session of the Kimberley Waterworks Company's plant on the Vaal 
River, some sixteen miles distant, and cut off the water-supply. Con- 
nections were made between Premier mine pumping system and the 
Kimberley Waterworks Company's reservoir, and a supply of eight to 




Railway Bridge over the Vaal River at Fourteen Streams, destroyed by the Boers. 

ten million gallons of water per month was delivered by De Beers, free 
of cost to that Company, on the understanding that only half rates 
should be charged to the inhabitants of the town. The water was per- 
fectly clear, pure, and wholesome. 

As the supplies of food in hand seemed ample for any emergency 
that was thought possible, there were practically no restrictions upon 
the consumption of supplies during the early part of the siege, except 
that the amount of meat was fixed at one pound per diem for each adult, 
and one-quarter of a pound for children under fifteen years of age. As 
there were no restrictions as to prices, the speculating part of the com- 
munity soon took advantage of the situation. Few had laid in stocks 
of food, and, as the greater number of people had not the means of mak- 
ing large purchases, they saw starvation staring them in the face. It 



APPENDIX I 



621 




Boer Laager, near Kimbedey. 

was impossible for many even to purchase their daily requirements at 
the fabulous prices to which the necessaries of life suddenly rose. Par- 
affin, which usually sold for 15 shillings a case, jumped to 100 shillings. 
Naturally the community rebelled against this extortion, and the daily news- 
paper was full of complaints. As some of them put it, they had taken 
up arms to defend the very people who were starving their families by 
putting the prices for the necessaries of life beyond their means. Colonel 
Kekewich was equal to the occasion, and wisely issued a proclamation 
fixing the price of all supplies at the same figures as formerly existed. 

For the support of people too poor to pay even for the barest neces- 
saries of life, thoughtful provision 
was made by Rhodes in the institu- 
tion of a soup kitchen in De Beers 
convict station. The details of the 
work were ably carried out under 
Captain Tyson, Dr. Smartt, and the 
Hon. Mrs. Maguire, the latter at- 
tending to the distribution at Bea- 
consfield. The soup was excellent, 
being composed of beef or horse- 
meat (with now and again a donkey 
or a few Angora goats thrown in), 
and a variety of vegetables from Group of Typical Boers. 

Kenilworth, and thickened with Boer meal or mealie meal. Captain 
Tyson carried pockets full of small bottles, the contents of which would 
be emptied in the brew, "just to make it a little more appetizing, don't 
you know." The allowance of meat was a half pound for two days, 




622 THE DIAMOND MINES OF SOUTH AFRICA 

which could be exchanged for soup. Long rows of people stood for 
hours awaiting their turn to be served. 

When the siege commenced, De Beers had 8000 tons of coal in 
stock and also about 2000 tons of wood. There were about 1500 
cases of dynamite belonging to merchants, and De Beers had several 
hundred cases in stock. Owing to the dangerous proximity of the mag- 
azine to the town, it became necessary to remove nearly all the dyna- 




A Group of Mercenaries, fighting with the Boers. 

mite to a magazine at Dronfield, about six miles north of Kimberley, 
from which, for a time, supplies were drawn ; but these magazines were 
subsequently blown up by the Boers. 

In order to do as much work as possible while the supply of coal and 
dynamite lasted, permission was obtained from the officer commanding 
for the miners to resume work in the mines, on condition that substi- 
tutes were found to take their places in the forts. A company of men 
was organized at De Beers and Kimberley mines by the assistant general 
manager, which was known during the siege as the Permanent Guard, and 
was composed mostly of refugees. Work was continued at Kimberley 
mine until the 3d of November, and at De Beers mine until the 4th of 
December, when it was thought advisable to discontinue work and save 
the supply of coal for pumping water for the use of the town and prevent- 
ing the mines from being flooded. The amount of ground hoisted at 
Kimberley mine from October 14 to November 3 was 60,396 loads, 
and at De Beers mine to December 4 was 173,447. The pumps in 



APPENDIX I 



623 



both mines were kept going until a few days before the siege was raised, 
and started again before the water had filled the tunnels in the rock out- 
side the mines proper. While the pumps were stopped a gang of natives 
were kept busy at each mine picking out pieces of coal from the old ash- 
heaps to supply the boilers with fuel. Fortunately all damage by flooding 
to the underground works was prevented. 

Communication was kept up between Kimberley and the nearest mil- 
itary post, which was at the Orange River bridge on the Kimberley-De 
Aar railway, by despatch riders who evaded the Boers and found shelter 
and remounts at several farms of friendly colonists. The distance was 



f 

' /4^/v<- f+t 



t-f .1 j rtAf 



TV" '" 'v r~*'* ".,"-,> 

UK t a, t f ^.'jttr', "'"'-y -t'-,. , 

<&*/& ,1 ,-.'"t""",7'f.^ f f S '. ' '' "^ 

; 
x ' . , ' u. .-i. 




Code Dispatches received during the Siege. 

eighty miles. Trooper Brown of the Cape Mounted Police carried the 
first despatch, and covered the distance in thirteen hours. Great credit is 
due to these men, who went to and fro at great peril to themselves. Fore- 
most among them were Brown, Cummings, Hambly, and Harding, but 
there were many others who did good work. The remuneration paid by 
the military was very small ^5 for the round trip, but in many cases, 
where private letters were carried, this sum was largely increased by pri- 
vate donations. Later, when the investment of the town was closer, it 
became very difficult to get through the Boer lines, and despatch riders, 
carrying private despatches, were paid as high as ;iOO for a round trip. 
Many of these men were captured and taken to Bloemfontein as prisoners 
of war. 



624 THE DIAMOND MINES OF SOUTH AFRICA 




How zealously and efficiently Rhodes took part in the preparations for 
the defence of Kimberley has been particularly noted by Mr. George A. 

L. Green, editor of the 
Diamond Fields Advertiser 
in his able and accurate 
description of the siege. 
" The need for mounted 
troops to watch the ene- 
my's movements was early 
felt. The formation of 
a new corps, to be called 
the Kimberley Light 
Horse, was one of the last 
things authorized by the 
High Commissioners be- 
fore Kimberley was cut 
off, but the trouble was 

Typical Boer. tQ fin( j {he horses Mr . 

Rhodes came to the rescue, and in a few days presented the corps with 
five hundred admirable mounts ; he also did some good work as recruit- 
ing sergeant. Largely through his 
efforts the mounted arm of the de- 
fence forces was thus increased to 
nearly nine hundred men. Major 
Scott-Turner was appointed with the 
local rank of lieutenant colonel to 
command the mounted corps, which 
now comprised Cape Mounted Police, 
Diamond Fields Horse, and Kimber- 
ley Light Horse. 

" It was Mr. Rhodes's pleasant 
custom to go round asking the ques- 
tion, ' Do you want anything ? ' 
Needless to say he rarely met any one 
who did not want something. 

" One evening, while Major Cha- Ma J r w " H " E " Murray ' 

mier was dining with Mr. Rhodes, they were discussing the artillery 
branch of the defence forces, when Mr. Rhodes asked him if he needed 
anything for his artillery. The Major replied quickly, ' Yes, I want 




APPENDIX I 



625 



to make my guns mobile. [Note. It is mentioned elsewhere that 
these guns were small mountain guns without limbers.] I require, to do 
that, 43 horses, 62 mules, 7 buck wagons, and 4 Scotch carts.' It was a 
tall order, but Mr. 
Rhodes made a men- 
tal note, without 
any comment, and 
three days later 
Major Chamier 
found that the 
whole requisition 
had been delivered 
at the artillery 
camp. All he could 
say, when he saw what had been done in so short a time, was, ' What 
a wonderful man Mr. Rhodes is.' It was an object lesson to the military 
officers to see how quickly provisions of this kind could be made by a 
civilian who was in no way handicapped by official red tape." 

From the first threat of the outbreak of hostilities, the resources of De 
Beers were at the command of the garrison for any needed service. At 




Armored Locomotive. 




Armored Train, constructed at De Beers Workshops. 

De Beers workshops several engines and trucks were armored in the 
manner shown in the accompanying illustrations. 

These trains were useful in many ways, and of very great service in 
keeping the lines of communication open. Those running between 
Kimberley and De Aar were manned in part by De Beers men. The 



2 s 



626 THE DIAMOND MINES OF SOUTH AFRICA 

military organization known as Scott's Railway Guards was also mostly 
made up of De Beers men, with Lieutenant Colonel R. G. Scott, one of 
the officers of De Beers Company, in charge. 

The first encounter with the attacking Boers was on the 24th of 
October, ten days after the investment of Kimberley. Shortly after the 
water-supply had been cut off, Lieutenant Colonel Scott-Turner made a 
reconnoissance in the direction of the pumping station, but took the pre- 
caution to follow the line of the railway as far as Macfarlane's farm, 
which lies eleven miles to the north of Kimberley. His force consisted of 
detachments of the Kimberley Light Horse, Captain R. G. Scott, V. C. ; 
Cape Mounted Police, Major Elliott ; and the Diamond Fields Horse, 
Major Rodger. The armored train, in charge of Lieutenant Webster, 




Railway Bridge at Modeler River, both ends of which were blown up by the Boers. 

Loyal North Lancashires, supported the troops. On arriving at the farm- 
houses at Macfarlane's, which stand on a knoll from which the country 
recedes in all directions, the troops halted and had breakfast. Immedi- 
ately afterward Lieutenant Colonel Scott-Turner with 1 80 men proceeded 
on his mission, but soon after his departure Boers were seen in several 
directions. 

Upon the appearance of the enemy Lieutenant Colonel Scott-Turner 
took up a strong position with his men. In a short time the Boers sent 
a few of their number under a flag of truce. Major Elliott of the Cape 
Mounted Police met them, and was told that if he and his command were 
on police duty the Boers would not molest them, but if he was there for 
a fight, they would put a bullet through his head. Major Elliott returned, 
however, without hindrance. In the meantime the armored train had 



APPENDIX I 



627 




Effect of a Nine-pound Shell. 

proceeded beyond Macfarlane's, but was soon recalled, as the Boers were 
evidently trying to cut it off. Later in the morning Boers continued to 
arrive from the north and east, and came within rifle range of Macfarlane's 
farm, not knowing 
that it was occu- 
pied by the Brit- 
ish. The patrol 
opened fire on them, 
and several of them 
were seen to fall 
and their riderless 
horses ran across 
the veld. The 
Boers retreated hel- 
ter-skelter. Shortly 
afterward five Boers 
from another com- Effect of a Nine-pound shell. 

mando came forward, bearing white flags, and were met by Major Elliott, 
who received the same message as before. The Boers evidently had 
little knowledge of the proper use of the white flag. 




628 THE DIAMOND MINES OF SOUTH AFRICA 




Trophies of the Siege The Author's Collection. 

In pursuing his advance Lieutenant Colonel Scott-Turner fell into 
an ambuscade, for, owing to the very long grass, which was nearly waist 
high, he was unable to detect the position of the Boers, who were strongly 
posted behind the wall of a dry reservoir in numbers greatly exceeding 
the British force. Not a shot was fired until the British came within 
easy rifle range, when they were met with such a fusillade from the maga- 
zine Mauser rifles that they sought the nearest cover. In this repulse the 
losses on the British side were three killed and nine wounded, and four- 
teen horses were killed or disabled. The wounded men were taken up 
and carried back with the retreating force, but the dead were left behind, 
to be brought in two days afterward, as the searching party failed to find 
the bodies, on the first day, in the tall grass. 

Lieutenant Colonel Scott-Turner had heliographed to the conning 
tower to have two mountain guns and two Maxims sent out. These 
were despatched at once, and the armored train took out 150 of the 
Loyal North Lancashires under command of Major Murray. The Boers 
were seen to be moving toward Dronfield, a ridge halfway between 



APPENDIX I 



629 



Kimberley and Macfarlane's. The armored train proceeded beyond 
Dronfield, but was ordered back to that place, and the troops left the 
train near the siding. In the meantime Captain May, with two guns, 
had reached a position just south of the siding, when the Boers opened 
fire on him at short range, having allowed his scouts to come close to the 
place where they were in ambush. Captain May quickly unlimbered his 
guns under a hot fire, and began to shell the Boers in return. Fortu- 
nately for him most of the Boer bullets went over the heads of his men, 
while he fired his guns with great precision, riddling the gamekeepers' 
houses, behind which the Boers had taken shelter, and soon driving them 
to the rocky ridge beyond. 

Hearing that the guns and regulars had 
gone out, I drove to a position north of Kenil- 
worth, where this part of the engagement was 
in full view. Captain May fired eighty rounds 
at the Boers, and his men behaved splendidly 
under a rain of bullets from the enemy, only 
a thousand yards distant. Out of a total of 
twenty-six men and eighteen horses, he had 
seven men wounded, three horses killed and 
nine wounded. Gunner Payne, who was 
wounded in the foot early in the fight, con- 
tinued to lay his gun until the end of the fir- 
ing ; and bugler Dickinson, who was wounded 
in the right hand, changed the bugle to his left 
hand and finished his notes. 

While this fight was going on Major Mur- 
ray had taken his men from the train near 
Dronfield, and had begun to ascend the hill. 
At his first advance the Boers opened fire. 
Forming his men in skirmishing order with all 
possible speed, he led the way up the rocky 
ridge where the Boers were lying closely under 
cover. Fortunately for the Major and his troops, the ascent of the Dron- 
field ridge on the north was comparatively easy, being over a gently rising 
country covered with small brush, with here and there a shallow ravine 
which gave a little shelter to his men. While they were moving for- 
ward, three men, not in uniform, rode up to him. At first he took them 
for Boers, but the Northumberland accent of the first who hailed him 




Boer 9 Ib. Shrapnel, % Actual 
Size, showing Time Fuse. 



630 THE DIAMOND MINES OF SOUTH AFRICA 

was convincing. They were men in charge of De Beers farms, and when 
the firing began they were looking after the large herd of De Beers cattle. 
One of these keepers, Dott, guided the troops up the hill, taking 
them out of sight of the enemy as much as possible, and shouting 
"This way, Mr. Officer!" "This way, Mr. Officer!" 




British and Boer Shells, fired within and around Kimberley and Beaconsfield. 

Their scramble up the hill was very plucky. In front lay the Boers 
hidden in the rocks, and on their left was a magazine containing 1500 
cases 37 y^ tons of dynamite, which might explode at any moment 
should a Boer bullet strike it, as it was protected only by a thin sheet of 
galvanized iron. Two firing parties of twenty-five each went ahead gallantly, 
with the main force, a hundred strong, following close behind. The men vol- 



APPENDIX I 631 

leyed and ran forward alternately, until they reached the crest of the hill, 
when they saw Commandant Botha and two or three companions standing 
near the large Griqualand West triangulation beacon which stood upon 
the summit. Most of the Boers made their escape by clambering over 
the precipitous ridge which forms the south and east boundary, but 
their brave commander, who held his ground to the last, was killed. 
The mass of Boers reached their horses, which stood among the trees 
below the ridge, and rode off pell-mell over the ridge in the distance, 
with shells from Captain May's guns bursting over them. 




Site of Cronje's I^aager, Magersfontein. 

In this engagement only one of the Boers beside Botha was killed, 
but seven were wounded. Major Murray had two officers and two men 
wounded. Colonel Scott-Turner and his men returned to Kimberley with- 
out meeting with any further opposition. 

It was fortunate that this reconnoissance was made, for the following 
despatch was taken from the body of Commandant Botha : 

" HOOF LAGER, OCTOBER 23, 1899. 
"VELDCORNET BOTHA, Bakinkop, Weledele Heer. 

" In reply to your inquiry about the taking of cattle in the neighbor- 
hood of Kenilworth, I am ordered by the Head Commandant Wessels 
to assure you that he considers it highly desirable that the same should 
be captured as soon as possible. 

" I am, &c. 

"J. B. M. HERTZOG." 

The success of this engagement was encouraging, but the fast-increas- 
ing numbers of the Boer besiegers and the extension of their lines soon 



632 THE DIAMOND MINES IN SOUTH AFRICA 



Colonel Kekewich, 
mandant Wessels's de- 
obtain possession 
vited to effect 
an operation 
the military 

The 






put a check on such excursions. Early in November Commandant 
Wessels offered to receive all Africander women and children into his 
own camp, and at the same time offered safe-conduct to all other women 
and children to the Orange River. The first part of his despatch was 
made public, but not the last. Wessels's despatch contained the following 
passage, " And whereas it is necessary for me to take possession of the 
town of Kimberley, i therefore I demand of your Honour that upon 
receipt of this you, as k^. Commanding Officer, shall forthwith 

hand over the town of flH Kimberley with all its troops and forts." 

in acknowledging receipt of Com- 
spatch, wrote, " Your desire being to 
of Kimberley, you are hereby in- 
the occupation of this town as 
\ of war by the employment of 
forces under your command." 
invitation was a challenge. 

On the morning of the 
6th of November, the Boers 
fired two shots at Premier 
mine, and on the following 
day the first actual bombard- 
ment began, from a position 
about five thousand yards 
from the mine. As the 
compound, containing over 

Conning Tower, De Beers No. I Shaft. twQ thousand nat i ve s, was 

close to the fort and in the direct line of fire, all these men were taken 
down into the open mine, where they were protected by an embankment 
150 feet high. 

On the same day other Boer guns commenced to bombard Kimberley 
from a ridge nearly five thousand yards distant. The British guns replied 
intermittently with a few shots. Kimberley had no ammunition to waste. 
The distances were so great that the little popguns in the Kimberley 
forts frequently " turned turtle," owing to the great elevation at which 
they had to be fired in order to carry the distance. The projectiles fell 
more like meteors out of the sky than shells from modern guns. For 
the first few days the Boer shells fell short of the inhabited part of Kim- 
berley. On the nth a shell burst in Dutoitspan Road, in front of the 
Catholic church, and killed an old Kafir woman, which was the only 




APPENDIX I 



633 




634 THE DIAMOND MINES OF SOUTH AFRICA 




Boer Trenches at Magersfontein. 



casualty from the two hundred shells fired into the town on that day. 

Seventy shells were fired by the Kimberley artillery during the day. 

The Kimberley mounted troops also engaged the Boers on the same 

day near Otto's Kopje mine, and troops under Major Peakman attacked 

the Boers on Car- 
ter's Ridge on 
their left flank. 

The cessation 
of active hostilities 
on Sunday made 
it a welcome day 
of rest to all the 
besieged, and no 
doubt to the be- 
siegers as well. 
It gave both sides 
the opportunity of 
praying long and 

hard that their enemies might be confounded. The first bombardment 

continued for five days, with no further serious casualties on the British 

side, and the townspeople, appalled at first, began to make light of the 

danger. More than half the shells fell without exploding, and many 

children as well as grown people ran 

up, after each shell struck, to carry 

off a trophy. These prizes and the 

fragments and fuses of exploded shells 

found ready purchasers. The military 

authorities issued an order forbidding 

people from collecting these shells 

and fragments, while a bombardment 

was going on, owing not only to the 

risk of death or maiming from the 

exploding shells, but to the greater 

danger of the explosion of the dyna- 

i_ i_ i j j Plan Through Tomb. 

mite mines which were laid around 

the town. The prohibiting order carried this warning, " These mines 
are at all times ' live,' that is, the fuses and firing arrangements are so 
arranged that the mines can be fired either automatically or by obser- 
vation, and they might under certain circumstances be ignited by the 




APPENDIX I 



635 



enemy's shells." This order should have frightened the average Kim- 
berley urchin, but its apparent effect was to make him all the more eager, 
for he seemed to think that he had a chance of finding a prize in one 
of those dynamite mines 
about which everybody was 
talking. 

As the siege dragged 
along, some of the Imperial 
officers began to grow impa- 
tient. Anticipating the ap- 
proach of Lord Methuen, 
they planned a sortie on the 
25th of November which 
was fairly successful ; for 
they took Carter's Ridge, 
some three miles to the west 
of Kimberley, and captured 
thirty-three Boers, including 
nine wounded. The fighting 
continued all day, and re- 
sulted in a loss to the garri- 
son of six killed and twenty- 
nine wounded, including Colonel Scott-Turner, and Captains Bowen 
and Hickson-Mahony. Towards evening the Kimberley troops re- 
turned to town, as their ammunition was giving out and it was getting 

too late to send for more. This was 
the first fight that many of these men 
had been in, and their gallantry was 
greatly creditable, though they were 
unable to hold the ground they had 
won. The Boers published their losses 
as nine killed, seventeen wounded, and 
fifteen missing, instead of thirty-three 
who were brought into Kimberley. 

On November a8th another attempt 
was made to drive the Boers from Car- 
Plan Through Columns. ter ' s Ridge. Shortly after noon there 
was great activity in town, and troops were moving in various directions 
making ready for a sortie. The centre of the advance, commanded by 




The Honoured Dead Memorial. 




636 THE DIAMOND MINES OF SOUTH AFRICA 

Colonel Scott-Turner, moved out in the direction of the reservoir and 
thence along a ridge which gave a little cover. The first Boer redoubts 
were quickly taken, and then Colonel Scott-Turner sent for two guns to 
support him. He drove the Boers back until they reached their last 
redoubt, a small fortress dug in the rock, with a coping of sand bags 
arranged with loopholes. Colonel Scott-Turner led his last charge and 
took cover in a small redoubt, only sixty yards from the Boers. There 
the Boers had their Armstrong gun. The Diamond Fields Artillery 
were obliged to cease firing, owing to the danger of shelling Colonel 
Scott-Turner and his little body of men. 




" Long Cecil" as a Mild Steel Billet. 

While this engagement was going on, a small troop of the Diamond 
Fields Horse attacked the Boer camp in the rear of their redoubt. This 
attack was successfully carried out by Captain Shackleton, who dealt the 
Boers a severe blow. He captured 149 loaded shells, a considerable 
quantity of gunpowder, a wagon and span (16 oxen), a Cape cart, and 
the limber of the gun which Colonel Scott-Turner was trying to take. 
Among the prizes was a baboon, which proved to be the mascot of the 
company of Cape Mounted Police stationed at Vryburg, left behind when 
they evacuated the town somewhat hurriedly. 

Meanwhile Scott-Turner and his men were in a most awkward posi- 
tion, lying in a shallow redoubt with its side partly exposed, for the re- 
doubts occupied by the two opposing parties both faced east toward 
Kimberley, but the one occupied by the Boers was much larger and bet- 



APPENDIX I 



63? 



ter built. It was impossible for any of the attacking party to show 
their heads without receiving a volley from the Boers, and thus one after 
another of these brave men fell back dead, until finally Scott-Turner 
took a rifle and was about to fire, when he fell, shot through the head. 
Major Peakman fought his way with a small force to one of the redoubts, 
within speaking distance of the survivors. Here he learned that Scott- 
Turner had been killed, and he at once assumed command as senior 
officer. He sent a message asking for reinforcements, but, before they 




"Long Cecil" in course of Construction. 

arrived, darkness had come on, and he decided to withdraw his men to 
Carter's farmhouse. On the following morning, ambulance wagons 
were sent out in charge of Captain Robertson under a flag of truce, to 
collect and bring in the dead. It was then ascertained that Kimberley 
had lost twenty-two killed and twenty-eight wounded, one of the latter 
being mortally hurt. 

In these encounters, as in all other occasions of their service during 
the siege, the ambulance corps was notably efficient, and the Kimberley 
doctors, as a body, did excellent service, both in the field under fire and 
in the hospitals. Particular mention may fitly be made of Drs. Heber- 
den and Ortlepp, who were attached to the mounted forces, and of Drs. 
Ashe, Mathias, McKenzie, and Watkins. 



638 THE DIAMOND MINES OF SOUTH AFRICA 

The fierceness of this engagement may be judged from Rhodes's state- 
ment at a De Beers meeting, held shortly after the siege, " I take this 
opportunity of placing it on record that seventy citizen soldiers of Kim- 
berley went to take the position, and out of that number there were only 
twenty who were able to creep away alive or unwounded after nightfall." 

The agth of November will long be remembered as the saddest day 
during the siege, when the brave men killed in this action were buried 
with military and civic honors. 

In order to meet the wants of the women and children whose bread- 
winners had fallen in battle, a fund was started ; to this De Beers gen- 




" Long Cecil," just before it was taken out of the Workshops. 

erously gave the sum of .10,000, and is now erecting a monument on 
one of the most elevated parts of the town, where the heroes who fell in 
the defence of Kimberley are to find their last resting-place. 

The object of these demonstrations was to detain as many of the be- 
sieging force as possible from leaving to join General Cronje at Modder 
River, and in this way to assist Lord Methuen in his advance to the 
relief of Kimberley. On December ist Lord Methuen's first search- 
light message reached Kimberley. This opening of communication was 
highly elating and all were eagerly expectant of the news. Word by 
word this message was spelled out, " Please inform the Remount Depart- 
ment, Wynburg, the number marked on the hoof of horse issued to Sur- 
geon O'Gorman of the Kimberley Garrison." 

Imagine the disappointment upon receiving this seemingly frivolous 



APPENDIX I 



639 



o 

UJ 

O 

03 

z. 
o 



o 

ID 

tr 

te 

z 
o 
o 




640 THE DIAMOND MINES OF SOUTH AFRICA 



LJI 









APPENDIX I 



641 



I 







2 T 



642 THE DIAMOND MINES OF SOUTH AFRICA 

message after the long ten weeks' investment. It was later reported that 
this communication was simply a test to ascertain whether the signals were 
passing between friends or enemies. 

On December nth Lord Methuen met with his first reverse in his 
march to Kimberley, where he was defeated by Cronje. I watched this 
battle from the conning tower, but, as the distance was about sixteen 
miles, one could see only the bursting of the shells, the big yellow cloud 
when a Lyddite shell exploded, and the captive balloon giving information 
as to the position of the Boers. One could hear the roar of the cannon, 
which sounded like the breaking of the sea against a cliff. We waited 




" Long Cecil " and the De Beers Men who made it. 

anxiously for news of the battle, but for days none came. The sus- 
pense was the more racking from the spread of the report that, as soon 
as Lord Methuen arrived, there would be an enforced exodus of all the 
women and children and male non-combatants from Kimberley. The 
carrying out of an order to this effect would inevitably have been attended, 
in my judgment, by great and needless suffering, and the reported deter- 
mination was rightly resented by all the citizens who had borne so pluckily 
the strain of the siege. 

At length, on December i8th, a week after the battle, we received 
the first authentic news that Lord Methuen had been defeated at Magers- 
fontein. This unlooked-for reverse, so blighting to sanguine hopes, cast 
a deep gloom over the beleaguered town, but there was no lack of heart 



APPENDIX I 



643 




" Long Cecil." 

in its stubborn defence. Christmas came, and with it the " Best wishes 
for Christmas Day and in the coming New Year," from the High Com- 
missioners, and also one from the Queen, " I wish you and all my brave 
soldiers a happy Christmas ; God protect and bless you all." These 
messages cheered the garrison and were given a most enthusiastic 
reception. 

On New Year's Day the mayor sent the following message on behalf 
of the citizens of Kimberley : " The inhabitants of Kimberley humbly 
beg to send your Majesty New Year's greeting. The troubles they have 




' Long Cecil " firing at the Boers on Carter's Ridge. 



644 THE DIAMOND MINES OF SOUTH AFRICA 

passed through, and are still enduring, only tend to intensify their love 
and loyalty towards your Majesty's throne and person ; " to which the 
Queen replied, " Am deeply touched by your kind message and New 
Year's greetings. I watch with admiration your determined and gallant 
defence, though I regret the unavoidable loss of life incurred." 

For some time after the repulse of Lord Methuen, siege life dragged on 
from day to day, with nothing very stirring to break the monotony. The 
various corps had their "At Homes," when tea would be served, and the 
Kimberley Regimental Band would enliven the throngs with martial music. 
Every little diversion from the dull routine of camp life was welcome. 




1. " Long Cecil," made in Kimberley. 

2. Royal Artillery Gun sent to defend Kimberley. 

To provide employment for as many of the inhabitants as possible, 
avenues were laid out and macadamized within the municipality of 
Kimberley and Beaconsfield, which add much to the convenience and 
beauty of the towns. In addition to this street work, Rhodes decided to 
make an avenue in commemoration of the siege and to be known as 
"Siege Avenue." Years before he had planted rows of grapevines 
ranging from 1000 to 2000 feet in length, which were trained upon 
trellises, but Siege Avenue was designed to outdo anything in the line of 
vine and tree planting that had been done in South Africa. Fourteen 
trenches, each over 6000 feet long, were dug. The two centre 
trenches were for vines and were 14 feet apart. There were trenches 



APPENDIX I 



645 




Casting Shells for Seven-pounders, De Beers Foundry. 




The Soup Kitchen. 



646 THE DIAMOND MINES OF SOUTH AFRICA 



on either side and at suitable distances for planting orange trees. The 
three outside trenches were for ornamental evergreen trees, such as 
the pepper, eucalyptus, Australian beefwood, and cypress, to serve as a 
protection to the vines and orange trees from the prevailing winds. 
Since the siege the vines and trees have been planted, and the wooden 
trellis has been erected, at a cost of nearly .3000. 

When the work of digging the trenches was first started, several hun- 
dred natives were employed. These trenches were about a mile from 
the nearest fort. As soon as the Y. A. O. (young artil- 
lery officer) in charge saw them, he telephoned about in 
these words to the O. C. in the conning tower; "A 
large party of Boers digging trenches just north of 
Kenilworth. Shall I open j^ fire on them ? " 
The reply came, "Wait ^l^k and ascertain if 
they are Boers." Y. A. J| A O. to O. C., " I 
don't think they are A ' ? Boers." A min- 







2 34 S 

i. Boer loo-Pounder. 2. Boer o-Pounder. 3. Boer Pom-Pom. 4. De Beers " Long Cecil," 
28-Pounder. 5- De Beers 7-Pounder. 

ute later Y. A. O. to O. C., " They are De Beers workmen digging 
trenches to plant trees." 

The old vines and fruit trees at Kenilworth were of incalculable 
value to the people of Kimberley, for they bore immense quantities of 
splendid fruit, which Rhodes sent to the hospitals, to the military camps, 
and to the citizens generally as far as it would go. 

In my own garden there must have been a ton and a half weight of 
beautiful grapes, which daily reminded one of the old saying, " It is more 
blessed to give than to receive," as one saw the look of joy on the faces 
of the women and children as they left the garden. My mulberry trees 
were also loaded with fruit, which was eagerly called for. Some substi- 
tute for butter or lard was particularly wanted, for neither of these was 



APPENDIX I 



647 




Notice on the Market Buildings during the Siege. 

procurable in the town. De Beers again was able to meet this call. In 
the great warehouses of the company were thousands upon thousands of 
gallons of lard oil kept in stock beautifully clear and sweet winter- 
strained lard oil. Hundreds of people came to the Company's stores 
daily for this supply. They fried their meat and bread in the oil, and 
found it much sweeter than most South African butter. 

In view of the now obvious certainty of the prolongation of the siege 
and the call for a gun of greater range and efficiency than any at the 
command of the garrison, the extraordinary task of the construction of the 
really formidable piece aptly named " Long Cecil " was undertaken by 
De Beers Company. It was designed by Mr. George Labram in De 
Beers workshops. Mr. Edward Goffe, chief draughtsman to the com- 
pany, describes the making of the gun expertly : 

" Long Cecil" was made from a mild steel billet, lof inches diam- 
eter and 10 feet long, weighing 2800 pounds, this being turned and 
rough bored to form the inner tube. 

The breech rings were forged from 6 inches x 2^ inches Lowmoor 
iron. They were turned and bored, and then shrunk on in place, nine 



648 THE DIAMOND MINES OF SOUTH AFRICA 

forming the first row shrunk on to the tube direct, and four more the 
second row over the breech, shrunken over the first row. The trunnion 
ring, carrying the trunnions or bearings, was forged in one piece without 
weld, from a length of 6 inches square Lowmoor iron, and was shrunk 
on against a shoulder left on the tube. The final boring was done 
after all the rings were shrunk on, the calibre of the gun being 4.1 
inches. The barrel was then ready for rifling. The rifling is a poly- 
groove increasing twist, consisting of 32 grooves, each ^ inch wide and 
JJT inch deep, which, starting with a pitch of I in 100 at the breech end, 




Inhabitants of Kimberley waiting (or their Daily Allowance of Four Ounces of Horse Meat. 

and increasing to a pitch of i in 32 in a length equal to 18 calibres 
73.8 inches, are uniform at that pitch for the remainder of the length. 
The curve of increase is the semi-cubical parabola. 

The breech block was made of mild steel, screwed to fit the breech 
with a " V " thread, flattened top and bottom, of | inch pitch. The 
" De Bange " system of obturation was adopted, the mushroom-headed 
bolt being of mild steel, annealed in melted tallow, and bored for the 
friction firing tube. The pad was made of rings of sheet asbestos 
soaked in melted tallow. 

The carriage was made of steel plates ^ inch thick, cut to shape, and 
riveted together in pairs, with gun-metal blocks between, for trunnion 
and axle bearings. The wheels were taken from a portable engine, 
bushed with gun metal and bored to fit the axle, whose ends were covered 
with brass dust-caps. 




APPENDIX I 



649 



The shells weighed 29 pounds each, loaded with their bursting charge 
of one pound of powder, and were fitted with the percussion fuse devised 
by Mr. George Labram. 

The making of the gun was begun on the 26th of December, 
1899. It was proved on the igth of January, 1900, and went into 
action January 23. From then up to the date of the relief of Kim- 
berly 255 shells were fired from the gun, mostly at ranges approximately 
5000 and 6000 yards, the distances of two of the positions of the 
enemy, which were easily reached with elevations of 12 and 15 re- 




Band of the Kimberley Volunteer Regiment playing at the Mounted Camp during the Siege. 

spectively, with powder charge of 5 Ibs. With the same powder charge 
another position 8010 yards distant was effectively reached with an ele- 
vation of 24 15'. 

The illustrations on previous pages show graphically how " Long 
Cecil " was made. 

The cut on page 643 represents the finished gun ready to go into 
action. On page 639, the upper figure shows the general construction 
of the gun barrel ; the lower figures on the same plate show the breech 
block and obturator. The rifling device is given on page 640, and on the 
same plate the boring and rifling tools are represented. 

The upper figure on page 641 gives the details of the construction 
of the shells and fuse for " Long Cecil," and underneath is shown the 
shell made for the 2. 5-inch popguns with which Kimberley was defended. 



650 THE DIAMOND MINES OF SOUTH AFRICA 




The Mafeking " Long Tom," manned by Mercenaries. 

The manufacture of these small shells was undertaken in November, and 
many were thrown into the Boer camps " with C. J. R.'s Compliments " 
stamped on them. 

The powder for charging these shells was fortunately at hand. The 
old Central Company of Kimberley mine had a large stock of black 
powder which was used for blasting in the open mine, as far back as 
1888. When De Beers Consolidated Mines took over this Company, 
the powder was removed and placed in a magazine on the veld, a mile 




" Long Tom," en route to Kimberley. 



APPENDIX I 



651 




The Boer loo-Pound 
Shell and the Imperial 
9-Pound Shell. 



beyond the Company's washing machines. Shortly after the opening of 
the siege I had stock taken of the contents of outlying magazines and 
brought to light three and a half tons of good black 
powder of various grain. This discovery was of 
much service, for it enabled the garrison to respond 
more frequently to the fire of the Boers, and made 
the construction and use of " Long Cecil " possible. 
At first, the shells cast in our foundry were not 
all perfect, and the bursting of some of them led to 
greater care in testing all under hydraulic pressure. 
Ring shells made by De Beers are shown on page 
641. Rings with jagged or saw-toothed edges were 
first cast ; these were stacked one over another 
in the mould, and the outer shell cast around them. 
When the bursting charge of powder exploded, these 
rings were broken into a hundred pieces and thrown 
in all directions. 

The Boers evidently resented the firing of" Long 
Cecil," for on the 24th of January they kept up a fierce cannonade, 
throwing about five hundred shells into Kimberley. A French officer, 
who was at Kampfersdam during a part of the siege, says 
that " Long Cecil " did good practice, and with one shell 
killed seven Boers, only two less than the Boers killed 
with eight thousand shells. The heavy and continuous 
firing which took place on the 25th of January and fol- 
lowing days caused many to build " shell-proofs " for the 
protection of the women and children. 
On the morning of the 8th of Feb- 
ruary, at about eleven o'clock, I was 
in the conning tower, and noticed an 
immense volume of smoke belched 
forth from a gun on Kampfersdam 
tailing heap. I remarked to those 
near me that the Boers had brought a 
" Long Tom " against us at last. In 
a few seconds the bang of the gun 
was heard, followed a little later by a sound almost indescribable as the 
shell came whizzing through the air. It has been likened not unfitly to 
the roar of an express train passing at full speed. Then a cloud of red 




Showing Interior of Boer loo-Pound Ring 
Shell and Shrapnel combined. 



652 THE DIAMOND MINES OF SOUTH AFRICA 



dust was seen where the shell had struck, shortly followed by the crash 
of the explosion. In the vicinity the air was filled with fragments of 
the shell or bullets of the shrapnel, which flew on with a singing " ping, 
ping, ping." Twenty-five of these shells were fired on that day, many 
of which did not explode. One was brought in and measured, and found 
to be fifteen centimetres, or about six inches, in diameter. 

The " Long Tom " 
which was brought to 
Kimberley was a cap- 
tured piece which had 
been struck by a shell 
on the muzzle and 
broken. This gun was 
taken to Pretoria or 
Johannesburg, where the 
broken part of the muz- 
zle was cut off and a 
band shrunk on the in- 
jured end. The illustra- 
tion of this gun on page 
650, on a railway truck 
en route from Pretoria 
to Bloemfontein, shows 
the method of moving 
these guns without a 
limber. The gun was 
noted for bad shooting. 
On the afternoon of 

Premier Studio, showing Effect of loo-Pound Shell. ^j^ ^ Q f February 

the Boers turned the gun on the herd of cattle which were being driven 
in for the night. This shot missed the cattle by half a mile to the left. 
Three more shots were fired, all falling wide of the target at which they 
were aimed. The illustrations here given of the effect of these shells 
are more graphic than words. 

On the first day the big gun was fired, the Buffalo Club was struck 
and sustained considerable damage, and a few private buildings were 
more or less injured. On the gth the firing of " Long Tom " com- 
menced at daybreak, and was continued at intervals throughout the day 
until six P.M., when 'the last shot was fired. This shot killed George 




APPENDIX I 



653 




Effect of a loo- Pounder. 




Effect of a loo-Pounder. 



654 THE DIAMOND MINES OF SOUTH AFRICA 




Labram, one of the most able men in the service of De Beers Company. 
He had entered his room in the Grand Hotel only a minute before. The 

shell passed through the roof 
and three brick walls before 
reaching Labram's room. Dur- 
ing the same day the wife and 
son (fifteen months old) of Mr. 
Robert Solomon were struck by 
the fragments of a shrapnel shell, 
which burst as it came through 
the outer wall of the building in 
which they were temporarily 
staying. The child was killed 
instantly, but the poor mother 
was taken to the hospital, where 
she died, thirty-six hours after- 
ward, from her injuries. 

During Saturday the firing 
continued, and buildings in every 
quarter of the town were struck. 
The peril of the unprotected 
people was appalling. There was the greatest activity in building shel- 
ters for the women and children. The tailing heaps were tunnelled, and 
the miners erected long rows of tunnel timbers against the debris em- 
bankments, and covered 
them with corrugated iron. 
Gangs of natives soon pro- 
tected these galleries with 
debris several feet deep. 
Still there were thousands 
unprovided with any shelter 
except the thin roofs and 
walls of their houses, which 
were absolutely useless 

against a hundred-pound Mr Compton . s Drawing-room barricaded for Shelter 

shell travelling at the rate of from Boer 9-Pounders. 

a thousand feet a second. When firing ceased, about midday, there was 
a sigh of relief from many hearts, for it was thought that firing would 
not be resumed until Monday morning. 



Effect of a loo-Pounder. 




APPENDIX I 



655 



The funeral of Mr. Labram was timed to leave the hospital at eight 
o'clock in the evening, as it was thought unsafe to have the funeral by day. 
He was buried with full military honors, and, as the hour for departure 
from the hospital approached, the streets were thronged with anxious and 
sorrowful people. The troops consisted of regulars, the various volunteer 
corps, and members of the Town Guard. My carriage contained Colonel 
Kekewich, Mr. Rhodes, Mr. Pickering, and myself. Other carriages 
followed, and hundreds who were unable to procure conveyances, owing 
to the scarcity of horses, joined on foot. 

Precisely at eight o'clock the procession moved from the hospital, but, 
before it had gone a hundred yards, the bugler in the conning tower 




Shell-proof, constructed by the Public Works Department. 

gave the well-known notes which meant that the big Boer gun had been 
fired. The band was playing the funeral march at the time, so that few 
people in the immediate vicinity of the hospital heard the warning notes. 
Shortly, however, the boom of the cannon was heard, followed by that 
never-to-be-forgotten hiss of the shell passing through the air. Traitors 
in the town had given the Boers information as to the time of the funeral, 
and doubtless signalled from some elevated place to the besiegers at 
Kampfersdam the moment the procession started. There was a sigh of 
relief as the fearful shell passed over the heads of the multitude, and fell 
harmless in vacant land behind the hospital. Colonel Kekewich gave 
orders for the band to cease playing, and that all carriage lights be put 
out. It was a grim and silent funeral. Shot after shot came thundering 



656 THE DIAMOND MINES OF SOUTH AFRICA 



over or into the town, as the procession passed through it. At last, as we 
approached the cemetery, we could see the flash of the gun as it was fired. 




Excavations in the Tailing Heaps at Beaconsfield, used as Shelters. 

While the last rites were said, the voice of the venerable archdeacon 
was drowned by the roar of the gun and the hissing of the shells. 

When the ceremony was over, every one hastened home to seek what- 
ever cover could be found. Crowds of people were massed for hours 

behind flimsy walls, 
which could not pro- 
tect them, but even 
this slight pretence 
of shelter was com- 
forting. The terrible 
night of the loth of 
February, 1900, will 
never be effaced from 
the memories of 
those who passed 
through it. 

So great was the 

Shell-proof, after the Siege. strain upon the 

nerves of the people that it was necessary that some one should come to 
their help, and as usual that "some one" was Rhodes. Early on Sunday 
morning he came to my house and said : " You told me, some time ago, 
that you could put a lot of people down in the mines, and I think the 




APPENDIX I 



657 



time has now come when we 
must do it. Will you get your 
mines ready so that the people 
can be sent down this evening?" 
I supervised the work at De 
Beers mine, and my son was its 
director at Kimberley. Tunnels 
one thousand and twelve hun- 
dred feet below the surface were 
cleaned out sanitary arrange- 
ments were provided, and, early 
in the afternoon, both mines were 
ready for occupation. Rhodes 
had sent a notice about town 
which is given as an illustration 

011 page 658, and Speaks for itself. Shell-proof at the Convent. 

Attention was called to it by the ringing of a bell. Crowds flocked to 
both shafts during the afternoon and evening ; and before midnight 
nearly three thousand women and children were safely housed, deep 
down in the subterranean passages of the mines. There was discom- 




Wf*mwjftf?W^*fi 

*' ''/ T/ ' '&ii' wi/jOi 

pR'/* iij ' y i i-'iH \ ^ i i-' i 

>' /" > ' ; ,!' '>)'< 



-. . - 



- ^- 




Shell-proof Dugouts. 

fort, of course, in this rude lodging, but all were happy in the thought 

that they were beyond the sound of screeching shells, and out of danger. 

I have never seen so much patience and pluck shown by women as 

was shown by those in the mines. There was no sign of fear in going 

2 u 



658 THE DIAMOND MINES OF SOUTH AFRICA 

down in the rough mine cages, and when they reached the station, they 
found to their joy that the tunnel was like a beautiful arcade, brilliantly 
lighted with electric rays. Food was served several times a day, and time 
went so quickly that dates were lost sight of and days and nights be- 
came hopelessly mixed. One lady asked me, " Is this yesterday or to-day 
or to-morrow ? " When the glad news was brought to them that Kim- 
berley was relieved, they scarcely believed it, and many preferred to re- 
main in the mines rather than take any chances of hearing " Long Tom " 




PUN DAY 

JANaCHILDREN VVHO.DESI BE 

I COM PLE TE.SH ELTE R.TO.PROCEi 
TO.Kl MBERLEXAN D.DE-BEEf 

SHAFTS. THEY.WILUBE.LOWEF 



8 O'CLCCH.THROUGHOUT.T 
LAMPS.&CU1DeS,WILL.B 

PROVIDED. r J R 









Notice sent round the Town during the Shelling by the Boers' loo-Pound Gun. 

give a parting roar, or the awful screech of a flying shell. On Friday 
morning all were brought to the surface, thankful for the few days of 
peace and safety. 

The illustration on page 660 shows the people at the shaft waiting to 
be sent down. That all were taken down into the mine and brought up 
again without the least mishap speaks well for those who carried out the 
details at each mine. 

The Boers fired a few shots between eleven and twelve o'clock on 
the 1 5th, from "Long Tom." They knew before we did that a British 
column was nearing Kimberley, for they had telegraphic communication 
between all their camps ; and while the column was slowly advancing they 
were using every effort to remove the big gun, which they did success- 



APPENDIX I 



659 




fully. Over eight thousand shells had been fired by the Boers into Kim- 
berley and its fortifications, with the result that, out of a total population 
of fifty thousand, only nine were killed, and the majority of these were 
women and children. 

At two P.M. a huge cloud of dust rose in the distant southeast, and 
shortly afterward one could see mounted troops advancing, and a helio- 
graphic message informed the officer commanding Kimberley that it was 
the relief column. The news spread like wildfire, and from every place 
which afforded a view, thousands of eager eyes were scanning the veld 




Too Late ! These two Siege Guns arrived after the Siege. 



660 THE DIAMOND MINES OF SOUTH AFRICA 




The United States Consulate. 




Women and Children waiting to be lowered down De Beers Mine. 



APPENDIX I 



66 1 



for a glimpse of the troops. The few public conveyances which were 
left in Kimberley were quickly taken to convey people to meet the 
column. 

As soon as I received the news, I made an effort to obtain a cab, but 
found it impossible. A small spring wagon drawn by a mule and driven 
by a Kafir passed my door at this time. Recognizing it as a De Beers 
fruit and vegetable wagon, I commandeered it, and in company with Cap- 
tain Bowen was driven to the Sanitorium, which afforded a good view of 




Double-decked Cage loaded with People and their Bedding being hoisted from the I2oo-foot 

level of De Beers Mine. 

the advancing troops. With my field glasses I saw the troops slowly 
advancing, and as they rounded a hill near the farmhouse on De Beers 
farm, Benaauwdheidsfontein, the Boer gun at Olefantsfontein, southeast 
of Premier mine, fired a few shots, but the relief column had a battery 
which soon silenced the Boer gun. Having telephoned for my light car- 
riage and horses, I soon joined the great crowd which thronged every 
road leading toward the advancing troops. It was seven o'clock before 
they got into camp. Thus Kimberley was relieved after a long and event- 
ful siege of 124 days. My old friend, Colonel Rimmington of Rimming- 
ton's Scouts, was among one of the first arrivals. Thinking that I was doing 



662 THE DIAMOND MINES OF SOUTH AFRICA 



him a good turn, I put him up at my house. I am sure he enjoyed the bath, 
but when I went to call him the next morning, at four o'clock, he was gone. 
Missing the bedclothes, search was made in the garden, and there the 
poor old tired soldier, wrapped up in clean sheets and blankets, was lying 
on the ground, sleeping as only a weary soldier can sleep. He had found 
the house too stuffy after sleeping so long on the veld. 

General French moved at daybreak the morning after his arrival, tak- 
ing with him about half of his column and four batteries of field guns. 
He gave battle to the Boers north of Kimberley, and cleared them out of 
their late haunts. The Boers left one gun behind, an old Armstrong 
gun, the limber of which was captured November 25th. On Saturday 

morning at daybreak General 
French left for Paardeberg, 
taking those of his troops who 
had rested on Friday, and the 
others followed the next day. 
It has often been asserted 
that Rhodes interfered with 
the military. He did suggest 
to Lord Methuen that there 
were more ways into Kim- 
berley than the one over the 
Magersfontein and Spytfontein 
kopjes, and mentioned the 
route over which General 
French came when he relieved 
Kimberley. He proposed that 




Boer Gun captured at Dronfield. 



small forts be built, every three or four miles, advancing from Modder 
River and keeping up the base of supplies at that place. His plan 
was substantially the blockhouse system, which the army later adopted, 
only that forts, instead of houses, would have been necessary, as the 
Boers then had cannon. The only reply to this suggestion was an order 
to the officers commanding Kimberley to have no communication what- 
ever with Mr. Rhodes on military subjects. 

Fortunately for the defence of Kimberley, Rhodes's energies were 
unflagging, in spite of rebuffs. Throughout the siege no appeal for assist- 
ance was ever made to him, nor even a want intimated on the part of the 
garrison, that he did not do all in his power to meet at once. The forma- 
tion of the Kimberley Light Horse was due to him. So, too, was the 



APPENDIX I 



663 



fortification of the village of Kenilworth and the outlying washing 
machines. The making of the gun, " Long Cecil," was by his order. 
The employment of thousands of idle hands in street-making in and 
around Kimberley was at his suggestion, and paid for by De Beers 
Company, thus assuring a support more welcome than charity. The 
undertaking of the soup kitchen was his proposal. From the great De 
Beers dairy milk was supplied to the hospitals, the sick at home, and to a 
depot where it was distributed under the supervision of a committee. 




Surrender of General Cronje. Reproduced by kind permission of Mr. L. S. Amery. 

Fruit and vegetables from De Beers gardens were sent to the hospitals, to 
the camps, and to the poorer families of the town. New gardens were 
started to enlarge the supply. The ice plant was kept constantly 
running, and ice furnished to the hospitals, the garrison, and the citizens 
generally. In everything contributing to the efficiency of the defence 
and the welfare of the people of Kimberley, Rhodes took the keenest 
interest, and, whenever possible, a most active part. 

A few days before the relief column came in, there was a meeting of a 
considerable number of the leading citizens of Kimberley, with the object 
of sending a message to Lord Roberts to inform him of the situation and 



664 THE DIAMOND MINES OF SOUTH AFRICA 

ascertain whether there was any immediate prospect of relief. Rhodes, 
the mayor and ex-mayor, a judge of the High Court, several members of 
Parliament, the author, and other citizens were present, and it was 
decided to send the following message to Lord Roberts, who was then at 




Laid Roberts and General Cronje seated under the Trees on the Modder River 
at Paardeberg. 

Modder River. The military censor at first refused to send it, but the 
officer commanding finally decided to permit its transmission in an 
abridged form. 

"KiMBERLEY, loth February. On behalf of the inhabitants of this 
town, we respectfully desire to be informed whether there is an intention 
on your part to make an immediate effort for our relief. Your troops have 
been for more than two months within a distance of a little over 20 
miles from Kimberley, and if the Spytfontein hills are too strong for 
them, there is an easy approach over a level flat. This town, with a 
population of over 45,000 people, has been besieged for 120 days, and 
a large portion of its inhabitants have been enduring great hardships. 
Scurvy is rampant among the natives ; children, owing to lack of proper 
food, are dying in great numbers, and dysentery and typhoid are very preva- 



APPENDIX I 665 

lent. The chief food of the whites has been bread and horseflesh for a 
long time past, and for the blacks meal and salt only. These hardships, 
we think you will agree, have been borne patiently and without complaint 
by the people. During the past few days the enemy has brought into 
action from a position within three miles of us a six-inch gun, throwing a 
hundred-pound shell which is setting fire to our buildings, and is daily 
causing death among the population. As you are aware, the military 
guns here are totally unable to cope with this new gun. The only 
weapon which gives any help is one locally manufactured. Under these 
circumstances, as representing this community, we feel that we are justi- 
fied in asking whether you have any immediate intention of instructing 
your troops to come to our relief. We understand that large reenforce- 




General Cronje as a Prisoner of War. 

ments have recently arrived at Cape Town, and we feel sure that your 
men at Modder River have, at the outside, 10,000 Boers opposed to them. 
You must be the judge as to what number of British troops would be 
required to deal with this body of men, but it is absolutely essential that 
immediate relief should be afforded to this place." 

The reply received from Lord Roberts was sent to Colonel Keke- 
wich, and was as follows : 

" I beg you will represent to the Mayor and Mr. Rhodes, as strongly 
as you possibly can, the disastrous and humiliating effect of surrender 
after so long and glorious defence. Many days cannot possibly elapse 
before Kimberley will be relieved, as we commence active operations 
to-morrow. Future military operations depend in a large degree on your 
maintaining your position a very short time longer." 



666 THE DIAMOND MINES OF SOUTH AFRICA 




Reception of Lord Roberts and Lord Kitchener, Kimberley Town Hall. 

What message or messages were sent by the military from Kimberley 
that conveyed to the mind of Lord Roberts that there was even the 
remotest chance of the citizens of Kimberley surrendering to the Boers, 
will probably always remain a military secret. Suffice it to say, however, 
that such a thought never entered the minds of the men of Kimberley, 
who would rather have died in their trenches than have surrendered, so 
long as any scrap of food remained. 

The saving of Kimberley from the attack of the Boers was due to 
the natural strength of the position and its improvised fortifications ; to 
the courage of the citizen soldiers, and the small force of Imperial 
troops ; to the indomitable spirit of Cecil John Rhodes, the chairman 
of De Beers Company, whose pent-up energies found vent in devising 
ways and means for adding to the plans of defence ; to the forethought 
of the De Beers men in charge of buying food for man and beast, who 
laid in supplies far in excess of any expected emergency ; and possibly, 
least of all, to the disinclination of the Boers to attack energetically a 
fortified town so long invested by commandoes greatly outnumbering the 
garrison under arms. 



APPENDIX II 

WINDING ENGINES FOR THE MAIN SHAFT, 
KIMBERLEY MINE 

THIS plant was designed by me for De Beers Consolidated Mines 
Limited, and built by Messrs. James Simpson & Co., of London, and 
consists of a pair of inverted vertical tandem compound-condensing 
engines, driving two reels, which are capable of carrying 1800 feet of 
flat ropes each. The principal dimensions of these engines are as 
follows, viz. : 

Diameter of high-pressure cylinders, two . . 1 9. 5 in. 
Diameter of low-pressure cylinders, two . 34 " 

Stroke of all cylinders . . . . 48 " 

Diameter of each air pump, two . . . 24 " 

Stroke of each air pump . . . .16" 

Diameter of steam cylinder of the reversing engine 7 " 
Diameter of oil cataract of the reversing engine . 4.75 " 

Stroke of reversing engine . . . .18" 

Diameter of each high-pressure steam pipe . . 6 " 

Diameter of each low-pressure steam pipe . . 10 " 
Diameter of each exhaust pipe to the condenser . 14 " 

Diameter of each high-pressure valve . . . 6 " 
Diameter of each low-pressure valve . . 8 " 

Diameter of each high-pressure piston rod . . 3.5 ' 

Diameter of each low-pressure piston rod . . 4.5 ' 

Diameter of crank-pins . . . . 5-5 ' 

Length of crank-pins . . . . 7 " 

Diameter of each main bearing . . . 14 " 

Length of each main bearing . . . . 32.5 " 

Diameter of main crank-shaft in the middle . 1 6 " 

Smallest diameter of each reel . . . . 9 ft. i-J- " 
Size of flat ropes used .... sfin. by^ " 
Capacity of blue ground skips . . . 100 cubic feet. 

667 



668 THE DIAMOND MINES OF SOUTH AFRICA 

These engines were intended to hoist six loads, each weighing 1600 
Ibs., from the lOOO-foot level in 45 seconds, including filling, starting, 
discharging, and stopping; but they do it in from 30 to 35 seconds. 

All the steam cylinders are fitted with the Corliss valve gear, having 
vacuum dash-pots, the cut-off" being effected by the same lever that works 
the throttle valves. 

Reversing is effected by ordinary links worked by eccentrics fitted on 
the tail-shafts ; the reversing engine being fitted with a floating lever so 
that the motion of the piston coincides exactly with the motion of the 
small hand lever. 

The two high pressure cylinders exhaust into the receiver, which is 
5 ft. diameter by 18 ft. long, fitted with sixty -eight 2-inch wrought- 
iron tubes, through which live steam from the high-pressure jackets, but 
at a reduced pressure, is constantly circulating. The object of this re- 
ceiver is to supply the low-pressure cylinders with a considerable volume 
of dry steam to facilitate a quick starting away. An 8-inch balanced 
throttle valve admits steam to the high-pressure cylinders, and a similar 
valve, 12 inches in diameter, admits steam from the reheater to the low- 
pressure cylinders. 

Each high-pressure cylinder is jacketed with live steam at full boiler 
pressure, the water of condensation together with a certain amount of 
steam passing through a Watts pressure regulator, which reduces the 
pressure in the jackets of the reheater and low-pressure cylinders to about 
30 Ibs. The final water of condensation is discharged automatically by 
a displacement trap into the hot well. 

Each air pump of the ordinary marine type is worked off the cross- 
head. The condenser, 6 ft. diameter by 16 ft. long, fitted with 125 
wrought-iron tubes 31^ in. outside diameter and 16 ft. long, is situated 
just outside the winding-engine house. All the water pumped from the 
mine passes through this condenser on its way to the floors. 

A circulating pump on the end of one of the tail-shafts supplies water 
for jet injection whenever the mine pumps are not supplying sufficient 
water to condense the steam. 

L. I. SEYMOUR, 

Mechanical Engineer for D. B. C. M. Ltd. 



APPENDIX III 

REPORT ON PUMPING PLANT FOR KIMBERLEY MINE 

THE new plant consists of a vertical triple-expansion condensing 
engine, having cylinders 15^ in., 23^ in., and 37 in. diameter respect- 
ively, with a stroke of 36 in. 

The high and intermediate pressure cylinders are arranged tandem, 
over one crank, the low pressure working on the other, which is placed 
at the opposite end of the crank-shaft and at an angle of 90 with the other. 

A double acting air pump is driven by a rocking lever from one cross- 
head and a feed pump in the same manner from the other engine. 

A cast steel spur-wheel, 3 ft. 9 in. pitch diameter, is keyed on the 
engine shaft, and drives a second shaft 27 in. diameter by gearing with a 
spur-wheel 30 ft. pitch diameter made of cast iron, with teeth 6 in. pitch 
by 30 in. face. The gears were made by Eraser & Chalmers, of Chi- 
cago, U.S.A., the crank-shafts by Sir J. Whitworth, of Manchester, and 
the rest of the work, including the pumps, by Messrs. J. Simpson & Co. 
Ltd., of London. A cast-steel crank is keyed on the second motion 
shaft, and drives the T bob by a pitman with 35 ft. centres. 

On the nose of the bob is hung the spear rod 1250 ft. long, of hard 
pine, 14 in. square for the first 500 ft., 12 in. square for the second, and 
10 in. square for the remainder. 

The total weight of the rod, including strapping plates and poles, is 
6 1 tons, which will be partially balanced by a counterweight on the top 
bob, and partly by a second bob placed at the i2OO-ft. level. 

Attached to the spear rod at the 250 ft., the 500 ft., the 750 ft., the 
1000 ft., and the 1200 ft. levels are cast-iron plungers 14 in. diameter, 
having a stroke of 10 ft., each of which forces the water to the next 
station above through a riveted steel pipe, 14 in. diameter, with joints 
riveted together. 

The foundations for the driving machinery are made of concrete, 
with the proportion of cement to stone of I : 9 on the average. 

L. I. SEYMOUR, Mechanical Engineer. 
669 



APPENDIX IV 

THE relative values of South African coals are shown in the following 
table, exhibiting tests made with the Beeley boilers at De Beers mine : 



ENGLISH COAL 

1890. 

April 29. Nixon's Steam Navigation coal 

July 1 6. Nixon's Steam Navigation coal (ist test) 

July 1 6. Nixon's Steam Navigation coal (zd test) 

Aug. 4. Nixon's Steam Navigation coal 



Pounds of feed-water 

evaporated per Ib. of coal 

from and at 212 Fahr. 

. 11.67 

. 10.22 

. I I. I I 

10.40 



SOUTH AFRICAN COAL 

Nov. 7. Vaal Drift Mine, Transvaal . 
1891. 

Feb. 12. Newcastle, Natal 

March 24. Indwe, Colonial .... 

March 25. Lewis and Marks, Transvaal . 

June 29. Newcastle, Natal 

July 20. Kroonstad, Free State . 



4-515 

9.520 
7.090 
6.734 
8.520 
7.084 



The relative cost and service of Welsh and Indwe coal delivered at 
the mines are approximately as follows : 

'- * 

A ton of 2000 pounds Welsh coal cost . . . 700 

A ton of Indwe coal . . . . . . 1190 

Welsh steam coal will evaporate about eleven pounds of water per 
pound of coal from and at 212 Fahr., and Indwe coal about seven 
pounds. Indwe coal is, therefore, worth about 6o|% of Welsh coal, 
and costs about .3 4*. for the same evaporating value contained in a ton 
of Welsh coal costing more than double this sum. 



670 



APPENDIX V 





O 


C 
B 
y 


cccccccc c c c 






S 1 

< w 


A 


SSiSi^^Si 10 *. . S. 2. o'5i>* Q 
(NCOMINOI-T't- 'f' ^J- iy Tf 




^ 

Q 
H 

H 




p 

Q S 

H 1 
s < 


y O 

- 2 
f 

en 


oooooooo o o 8 S S 
ooioinvniooo o O">ca.S O 

MMl-cOOOOOMN M* 1 !. 

ooONencnenenoooo oo oo o?- oo 

vOtx M NNWVOlO If) ^Ov'i' "> 

M" M M" M M" t-T 


I 

g 
| 


!3 


sduitrj jo 


1 


MooONomm>-"Mtxi-i moo eno^M M NO 
cocooomrnoenONtxONO t^o inen^ ON 




CO 

H 


IB sjoo|j uo anjg 

JO SpBO r J JO I,i|llill v^ 




vOiO*f"-OoOONO^ M txts,txtxP) QNO en 
t>.NMOO\OOrxCxCl enON *1-C* ON 

M'nTt-J'cfenMWof of efM'efw'oi' M" 


^ 


g 1 


M C